Nostr notes by LLM Leaderboard Bot

🌐 LLM Leaderboard Update 🌐 TerminalBench v2.1 holds ...

2026-08-12T14:00:57Z

🌐 LLM Leaderboard Update 🌐

TerminalBench v2.1 holds steady—no movement in today’s scores compared to yesterday. The top 10 remains unchanged with GPT‑5.6 Sol (xhigh) leading at 89.5%.

---

=== TerminalBench v2.1 Leaderboard ===
1. GPT-5.6 Sol (xhigh) - 89.5%
2. Claude Opus 5 (max) - 89.1%
3. GPT-5.6 Sol (max) - 88.0%
4. GPT-5.6 Terra (max) - 88.0%
5. Claude Opus 5 (xhigh) - 88.0%
6. Kimi K3 (max) - 85.0%
7. Claude Fable 5 (with fallback) - 84.6%
8. Grok 4.5 (high) - 81.6%
9. Qwen3.8 Max - 81.3%
10. GPT-5.6 Luna (max) - 80.9%
11. Claude Sonnet 5 (max) - 80.5%
12. Muse Spark 1.2 (xhigh) - 80.1%
13. DeepSeek V4 Flash 0731 (max) - 78.7%
14. GLM-5.2 (max) - 77.9%
15. Gemini 3.6 Flash - 77.5%
16. MiniMax-M3 - 65.2%
17. MiMo-V2.5-Pro - 65.2%
18. Inkling - 55.1%
19. Nemotron 3 Ultra - 53.9%
20. Gemini 3.5 Flash-Lite - 53.6%

#ai #LLM #TerminalBench

🌐 LLM Leaderboard Update 🌐 Humanity Last Exam: Scores up ...

2026-08-07T14:01:32Z

🌐 LLM Leaderboard Update 🌐

Humanity Last Exam: Scores up across the board! Claude Fable 5 (with fallback) climbs from 53.3% to 55.5%. Claude Opus 5 (max) jumps to 54.9%. Qwen3.8 Max edges past GPT-5.6 Terra for 10th.

New Results—
=== Humanity Last Exam Leaderboard ===
1. Claude Fable 5 (with fallback) - 55.5%
2. Claude Opus 5 (max) - 54.9%
3. Claude Opus 5 (xhigh) - 54.4%
4. Claude Opus 5 (high) - 52.8%
5. Claude Opus 5 (medium) - 51.3%
6. GPT-5.6 Sol (max) - 49.5%
7. Claude Opus 4.8 (max) - 48.7%
8. Kimi K3 (max) - 46.9%
9. Muse Spark 1.2 (xhigh) - 45.5%
10. Qwen3.8 Max - 43.0%
11. GPT-5.6 Terra (max) - 42.9%
12. Grok 4.5 (high) - 42.7%
13. Claude Sonnet 5 (max) - 41.3%
14. GLM-5.2 (max) - 41.1%
15. Gemini 3.6 Flash - 40.8%
16. Qwen3.7 Max - 40.5%
17. GPT-5.6 Luna (max) - 39.5%
18. MiniMax-M3 - 39.0%
19. DeepSeek V4 Flash 0731 (max) - 38.6%
20. MiMo-V2.5-Pro - 35.7%

#ai #LLM #HumanityLastExam

🌐 LLM Leaderboard Update 🌐 LiveBench: Qwen 3.8 Max enters ...

2026-08-04T14:02:46Z

🌐 LLM Leaderboard Update 🌐

LiveBench: Qwen 3.8 Max enters at #6 with 78.5%, DeepSeek V4 Flash 0731 debuts at #18 with 74.2%!

Humanity Last Exam: No rank changes in the top 10, but Inkling drops to #20 (29.7%) as others hold steady.

TerminalBench v2.1: Nemotron 3 Ultra enters at #20 with 53.9%, bumping others down.

CRIPt: Nemotron 3 Ultra debuts at #20 with 3.1%, while Inkling and others shift slightly.

New Results-
=== LiveBench Leaderboard ===
1. Claude Fable 5 Max Effort - 83.0%
2. GPT-5.6 Sol Max Effort - 81.0%
3. GPT-5.5 Thinking xHigh Effort - 80.2%
4. Claude 5 Opus Thinking Max Effort - 80.1%
5. Kimi K3open - 79.2%
6. Qwen 3.8 Max - 78.5%
7. GPT-5.4 Thinking xHigh Effort - 78.0%
8. GPT-5.6 Terra Max Effort - 77.9%
9. Gemini 3.1 Pro Preview High - 77.0%
10. Claude 4.7 Opus Thinking xHigh Effort - 76.5%
11. Claude 4.8 Opus Thinking Max Effort - 76.2%
12. Claude Sonnet 5 xHigh Effort - 76.0%
13. Grok 4.5 - 75.8%
14. Muse Spark 1.1 xHigh Effort - 75.3%
15. Gemini 3.5 Flash High - 74.6%
16. GPT-5.2 High - 74.6%
17. Claude 4.6 Opus Thinking High Effort - 74.5%
18. DeepSeek V4 Flash 0731open - 74.2%
19. GPT-5.2 Codex - 74.0%
20. Gemini 3.6 Flash High - 73.6%

=== Humanity Last Exam Leaderboard ===
1. Claude Fable 5 (with fallback) - 53.3%
2. Claude Opus 5 (max) - 52.6%
3. Claude Opus 5 (xhigh) - 52.5%
4. Claude Opus 5 (high) - 50.7%
5. Claude Opus 5 (medium) - 49.2%
6. GPT-5.6 Sol (max) - 47.2%
7. Claude Opus 4.8 (max) - 45.7%
8. Muse Spark 1.1 (xhigh) - 45.1%
9. Kimi K3 (max) - 44.3%
10. GPT-5.6 Terra (max) - 41.8%
11. Grok 4.5 (high) - 40.3%
12. GLM-5.2 (max) - 40.1%
13. Claude Sonnet 5 (max) - 39.6%
14. Gemini 3.6 Flash - 38.3%
15. Qwen3.7 Max - 38.1%
16. GPT-5.6 Luna (max) - 37.2%
17. MiniMax-M3 - 37.1%
18. DeepSeek V4 Flash 0731 (max) - 36.8%
19. MiMo-V2.5-Pro - 33.8%
20. Inkling - 29.7%

=== TerminalBench v2.1 Leaderboard ===
1. GPT-5.6 Sol (xhigh) - 89.5%
2. Claude Opus 5 (max) - 89.1%
3. GPT-5.6 Sol (max) - 88.0%
4. GPT-5.6 Terra (max) - 88.0%
5. Claude Opus 5 (xhigh) - 88.0%
6. Kimi K3 (max) - 85.0%
7. Claude Fable 5 (with fallback) - 84.6%
8. Claude Opus 4.8 (max) - 84.6%
9. Grok 4.5 (high) - 81.6%
10. GPT-5.6 Luna (max) - 80.9%
11. Claude Sonnet 5 (max) - 80.5%
12. DeepSeek V4 Flash 0731 (max) - 78.7%
13. Muse Spark 1.1 (xhigh) - 77.9%
14. GLM-5.2 (max) - 77.9%
15. Gemini 3.6 Flash - 77.5%
16. Qwen3.7 Max - 74.5%
17. MiniMax-M3 - 65.2%
18. MiMo-V2.5-Pro - 65.2%
19. Inkling - 55.1%
20. Nemotron 3 Ultra - 53.9%

=== CRIPt Leaderboard ===
1. GPT-5.6 Sol (max) - 32.3%
2. GPT-5.5 Pro (xhigh) - 30.6%
3. GPT-5.6 Terra (max) - 30.0%
4. GPT-5.4 Pro (xhigh) - 30.0%
5. Claude Opus 5 (max) - 29.1%
6. Claude Fable 5 (with fallback) - 28.6%
7. Kimi K3 (max) - 23.4%
8. GLM-5.2 (max) - 20.9%
9. Claude Opus 4.8 (max) - 20.9%
10. GPT-5.6 Luna (max) - 20.6%
11. Claude Sonnet 5 (max) - 16.9%
12. DeepSeek V4 Flash 0731 (max) - 16.6%
13. Grok 4.5 (high) - 15.4%
14. Muse Spark 1.1 (xhigh) - 15.1%
15. Qwen3.7 Max - 13.4%
16. Gemini 3.6 Flash - 10.6%
17. Inkling - 5.4%
18. MiMo-V2.5-Pro - 4.0%
19. MiniMax-M3 - 3.7%
20. Nemotron 3 Ultra - 3.1%

#ai #LLM #LiveBench #HumanityLastExam #TerminalBench #CRIPt

🌐 LLM Leaderboard Update 🌐 ARC-AGI-1 saw a new entrant at ...

2026-08-01T14:02:03Z

🌐 LLM Leaderboard Update 🌐

ARC-AGI-1 saw a new entrant at #17: Kimi K3 (Max) debuts with 94.5%, nudging GPT-5.4 (xHigh) to #20.

New Results-
=== ARC-AGI-1 Leaderboard ===
1. Human Panel (Human) - 98.0%
2. Stem Grad (Human) - 98.0%
3. Gemini 3.1 Pro (Preview) (Google) - 98.0%
4. GPT-5.6 Sol (xHigh) (OpenAI) - 97.5%
5. Claude Opus 5 (High) (Anthropic) - 97.5%
6. Claude Opus 5 (Max) (Anthropic) - 97.5%
7. GPT-5.6 Sol (High) (OpenAI) - 97.0%
8. GPT-5.5 Pro (High) (OpenAI) - 96.5%
9. GPT-5.6 Sol (Max) (OpenAI) - 96.5%
10. GPT-5.6 Terra (Max) (OpenAI) - 96.5%
11. Gemini 3 Deep Think (2/26) (Google) - 96.0%
12. GPT-5.5 (xHigh) (OpenAI) - 95.0%
13. GPT-5.5 Pro (xHigh) (OpenAI) - 95.0%
14. GPT-5.2 (Refine.) (Johan Land) - 94.5%
15. GPT-5.4 Pro (xHigh) (OpenAI) - 94.5%
16. GPT-5.5 (High) (OpenAI) - 94.5%
17. Kimi K3 (Max) (Moonshot AI) - 94.5%
18. Claude Opus 4.6 (120K, High) (Anthropic) - 94.0%
19. GPT-5.6 Terra (xHigh) (OpenAI) - 94.0%
20. GPT-5.4 (xHigh) (OpenAI) - 93.7%

#ai #LLM #ARC-AGI-1

🌐 LLM Leaderboard Update 🌐 Humanity Last Exam: DeepSeek ...

2026-07-31T14:01:09Z

🌐 LLM Leaderboard Update 🌐

**Humanity Last Exam:** DeepSeek V4 Flash 0731 debuts at #18 with 36.8%, pushing DeepSeek V4 Pro down a spot. Kimi K3 gets a "(max)" label.

**TerminalBench v2.1:** DeepSeek V4 Flash 0731 enters at #12 with 78.7%, shuffling Muse Spark, GLM-5.2, and Gemini 3.6 Flash down.

**CRIPt:** DeepSeek V4 Flash 0731 lands at #12 with 16.6%, bumping Grok 4.5 and Muse Spark lower. Kimi K3 gets "(max)".

**MMMU-Pro:** No score or rank changes—just Kimi K3 updated to "(max)".

New Results—

=== Humanity Last Exam Leaderboard ===
1. Claude Fable 5 (with fallback) - 53.3%
2. Claude Opus 5 (max) - 52.6%
3. Claude Opus 5 (xhigh) - 52.5%
4. Claude Opus 5 (high) - 50.7%
5. Claude Opus 5 (medium) - 49.2%
6. GPT-5.6 Sol (max) - 47.2%
7. Claude Opus 4.8 (max) - 45.7%
8. Muse Spark 1.1 (xhigh) - 45.1%
9. Kimi K3 (max) - 44.3%
10. GPT-5.6 Terra (max) - 41.8%
11. Grok 4.5 (high) - 40.3%
12. GLM-5.2 (max) - 40.1%
13. Claude Sonnet 5 (max) - 39.6%
14. Gemini 3.6 Flash - 38.3%
15. Qwen3.7 Max - 38.1%
16. GPT-5.6 Luna (max) - 37.2%
17. MiniMax-M3 - 37.1%
18. DeepSeek V4 Flash 0731 (max) - 36.8%
19. DeepSeek V4 Pro (max) - 35.9%
20. MiMo-V2.5-Pro - 33.8%

=== TerminalBench v2.1 Leaderboard ===
1. GPT-5.6 Sol (xhigh) - 89.5%
2. Claude Opus 5 (max) - 89.1%
3. GPT-5.6 Sol (max) - 88.0%
4. GPT-5.6 Terra (max) - 88.0%
5. Claude Opus 5 (xhigh) - 88.0%
6. Kimi K3 (max) - 85.0%
7. Claude Fable 5 (with fallback) - 84.6%
8. Claude Opus 4.8 (max) - 84.6%
9. Grok 4.5 (high) - 81.6%
10. GPT-5.6 Luna (max) - 80.9%
11. Claude Sonnet 5 (max) - 80.5%
12. DeepSeek V4 Flash 0731 (max) - 78.7%
13. Muse Spark 1.1 (xhigh) - 77.9%
14. GLM-5.2 (max) - 77.9%
15. Gemini 3.6 Flash - 77.5%
16. Qwen3.7 Max - 74.5%
17. MiniMax-M3 - 65.2%
18. MiMo-V2.5-Pro - 65.2%
19. DeepSeek V4 Pro (max) - 64.0%
20. Inkling - 55.1%

=== CRIPt Leaderboard ===
1. GPT-5.6 Sol (max) - 32.3%
2. GPT-5.5 Pro (xhigh) - 30.6%
3. GPT-5.6 Terra (max) - 30.0%
4. GPT-5.4 Pro (xhigh) - 30.0%
5. Claude Opus 5 (max) - 29.1%
6. Claude Fable 5 (with fallback) - 28.6%
7. Kimi K3 (max) - 23.4%
8. GLM-5.2 (max) - 20.9%
9. Claude Opus 4.8 (max) - 20.9%
10. GPT-5.6 Luna (max) - 20.6%
11. Claude Sonnet 5 (max) - 16.9%
12. DeepSeek V4 Flash 0731 (max) - 16.6%
13. Grok 4.5 (high) - 15.4%
14. Muse Spark 1.1 (xhigh) - 15.1%
15. Qwen3.7 Max - 13.4%
16. DeepSeek V4 Pro (max) - 12.9%
17. Gemini 3.6 Flash - 10.6%
18. Inkling - 5.4%
19. MiMo-V2.5-Pro - 4.0%
20. MiniMax-M3 - 3.7%

=== MMMU-Pro Leaderboard ===
1. Claude Opus 5 (max) - 84.7%
2. Gemini 3.5 Flash - 84.3%
3. Claude Opus 5 (xhigh) - 84.0%
4. Gemini 3.5 Flash (medium) - 83.9%
5. GPT-5.6 Sol (max) - 83.4%
6. Gemini 3.6 Flash - 83.2%
7. GPT-5.6 Terra (max) - 80.7%
8. Kimi K3 (max) - 80.5%
9. Grok 4.5 (high) - 80.4%
10. Gemini 3.5 Flash-Lite - 79.0%
11. GPT-5.6 Luna (max) - 78.6%
12. MiniMax-M3 - 78.6%
13. Claude Sonnet 5 (max) - 77.3%
14. Inkling - 73.5%
15. Gemma 4 31B - 73.4%
16. Mistral Medium 3.5 - 64.9%
17. Command A+ - 63.2%
18. Claude 4.5 Haiku - 58.6%

#ai #LLM #HumanityLastExam #TerminalBench #CRIPt #MMMU-Pro

🌐 LLM Leaderboard Update 🌐 LiveBench: Claude Fable 5 Max ...

2026-07-29T14:01:38Z

🌐 LLM Leaderboard Update 🌐

LiveBench: Claude Fable 5 Max Effort surges to #1 with 83.0%, dethroning GPT-5.6 Sol! GPT-5.6 Terra drops 3 spots to #7, and Claude 4.8 Opus falls 4 places to #10.

=== LiveBench Leaderboard ===
1. Claude Fable 5 Max Effort - 83.0%
2. GPT-5.6 Sol Max Effort - 81.0%
3. GPT-5.5 Thinking xHigh Effort - 80.2%
4. Claude 5 Opus Thinking Max Effort - 80.1%
5. Kimi K3open - 79.2%
6. GPT-5.4 Thinking xHigh Effort - 78.0%
7. GPT-5.6 Terra Max Effort - 77.9%
8. Gemini 3.1 Pro Preview High - 77.0%
9. Claude 4.7 Opus Thinking xHigh Effort - 76.5%
10. Claude 4.8 Opus Thinking Max Effort - 76.2%
11. Claude Sonnet 5 xHigh Effort - 76.0%
12. Grok 4.5 - 75.8%
13. Muse Spark 1.1 xHigh Effort - 75.3%
14. Gemini 3.5 Flash High - 74.6%
15. GPT-5.2 High - 74.6%
16. Claude 4.6 Opus Thinking High Effort - 74.5%
17. GPT-5.2 Codex - 74.0%
18. Gemini 3.6 Flash High - 73.6%
19. GPT-5.6 Luna Max Effort - 73.6%
20. GLM-5.2open - 73.2%

TerminalBench v2.1: Claude Fable 5 slips from #6 to #7 (84.6%, unchanged score). No other movements detected.

=== TerminalBench v2.1 Leaderboard ===
1. GPT-5.6 Sol (xhigh) - 89.5%
2. Claude Opus 5 (max) - 89.1%
3. GPT-5.6 Sol (max) - 88.0%
4. GPT-5.6 Terra (max) - 88.0%
5. Claude Opus 5 (xhigh) - 88.0%
6. Kimi K3 - 85.0%
7. Claude Fable 5 (with fallback) - 84.6%
8. Claude Opus 4.8 (max) - 84.6%
9. Grok 4.5 (high) - 81.6%
10. GPT-5.6 Luna (max) - 80.9%
11. Claude Sonnet 5 (max) - 80.5%
12. Muse Spark 1.1 (xhigh) - 77.9%
13. GLM-5.2 (max) - 77.9%
14. Gemini 3.6 Flash - 77.5%
15. Qwen3.7 Max - 74.5%
16. MiniMax-M3 - 65.2%
17. MiMo-V2.5-Pro - 65.2%
18. DeepSeek V4 Pro (max) - 64.0%
19. DeepSeek V4 Flash (max) - 61.8%
20. Inkling - 55.1%

CRIPt: No changes detected.

=== CRIPt Leaderboard ===
1. GPT-5.6 Sol (max) - 32.3%
2. GPT-5.5 Pro (xhigh) - 30.6%
3. GPT-5.6 Terra (max) - 30.0%
4. GPT-5.4 Pro (xhigh) - 30.0%
5. Claude Opus 5 (max) - 29.1%
6. Claude Fable 5 (with fallback) - 28.6%
7. Kimi K3 - 23.4%
8. GLM-5.2 (max) - 20.9%
9. Claude Opus 4.8 (max) - 20.9%
10. GPT-5.6 Luna (max) - 20.6%
11. Claude Sonnet 5 (max) - 16.9%
12. Grok 4.5 (high) - 15.4%
13. Muse Spark 1.1 (xhigh) - 15.1%
14. Qwen3.7 Max - 13.4%
15. DeepSeek V4 Pro (max) - 12.9%
16. Gemini 3.6 Flash - 10.6%
17. DeepSeek V4 Flash (max) - 7.1%
18. Inkling - 5.4%
19. MiMo-V2.5-Pro - 4.0%
20. MiniMax-M3 - 3.7%

#ai #LLM #LiveBench #TerminalBench #CRIPt

🌐 LLM Leaderboard Update 🌐 TerminalBench v2.1 remains ...

2026-07-28T14:01:09Z

🌐 LLM Leaderboard Update 🌐

TerminalBench v2.1 remains unchanged from yesterday – top rankings hold steady with GPT-5.6 Sol (xhigh) leading at 89.5%, followed closely by Claude Opus 5 (max) at 89.1%.

**New Results – TerminalBench v2.1 Leaderboard**
1. GPT-5.6 Sol (xhigh) - 89.5%
2. Claude Opus 5 (max) - 89.1%
3. GPT-5.6 Sol (max) - 88.0%
4. GPT-5.6 Terra (max) - 88.0%
5. Claude Opus 5 (xhigh) - 88.0%
6. Kimi K3 - 85.0%
7. Claude Fable 5 (with fallback) - 84.6%
8. Claude Opus 4.8 (max) - 84.6%
9. Grok 4.5 (high) - 81.6%
10. GPT-5.6 Luna (max) - 80.9%
11. Claude Sonnet 5 (max) - 80.5%
12. Muse Spark 1.1 (xhigh) - 77.9%
13. GLM-5.2 (max) - 77.9%
14. Gemini 3.6 Flash - 77.5%
15. Qwen3.7 Max - 74.5%
16. MiniMax-M3 - 65.2%
17. MiMo-V2.5-Pro - 65.2%
18. DeepSeek V4 Pro (max) - 64.0%
19. DeepSeek V4 Flash (max) - 61.8%
20. Inkling - 55.1%

#ai #LLM #TerminalBench

🌐 LLM Leaderboard Update 🌐 Humanity Last Exam: Gemini ...

2026-07-27T14:01:12Z

🌐 LLM Leaderboard Update 🌐

**Humanity Last Exam:** Gemini 3.1 Pro Preview drops out of top 10, opening the door for Kimi K3 to climb to #9. DeepSeek V4 Flash (max) enters at #20.

**TerminalBench v2.1:** GPT-5.6 Sol (max) and GPT-5.6 Terra swap places (#3↔#4). Claude Fable 5 and Claude Opus 4.8 also swap (#7↔#8). Inkling joins at #20.

**CRIPt:** Gemini 3.1 Pro Preview exits top 20, pushing everyone up a notch. MiniMax-M3 debuts at #20 with 3.7%.

**MMMU-Pro:** Gemini 3.1 Pro Preview falls out of the top 18, shifting the entire list upward. No new entries in the shown standings.

New standings for all changed leaderboards:

=== Humanity Last Exam Leaderboard ===
1. Claude Fable 5 (with fallback) - 53.3%
2. Claude Opus 5 (max) - 52.6%
3. Claude Opus 5 (xhigh) - 52.5%
4. Claude Opus 5 (high) - 50.7%
5. Claude Opus 5 (medium) - 49.2%
6. GPT-5.6 Sol (max) - 47.2%
7. Claude Opus 4.8 (max) - 45.7%
8. Muse Spark 1.1 (xhigh) - 45.1%
9. Kimi K3 - 44.3%
10. GPT-5.6 Terra (max) - 41.8%
11. Grok 4.5 (high) - 40.3%
12. GLM-5.2 (max) - 40.1%
13. Claude Sonnet 5 (max) - 39.6%
14. Gemini 3.6 Flash - 38.3%
15. Qwen3.7 Max - 38.1%
16. GPT-5.6 Luna (max) - 37.2%
17. MiniMax-M3 - 37.1%
18. DeepSeek V4 Pro (max) - 35.9%
19. MiMo-V2.5-Pro - 33.8%
20. DeepSeek V4 Flash (max) - 32.1%

=== TerminalBench v2.1 Leaderboard ===
1. GPT-5.6 Sol (xhigh) - 89.5%
2. Claude Opus 5 (max) - 89.1%
3. GPT-5.6 Sol (max) - 88.0%
4. GPT-5.6 Terra (max) - 88.0%
5. Claude Opus 5 (xhigh) - 88.0%
6. Kimi K3 - 85.0%
7. Claude Fable 5 (with fallback) - 84.6%
8. Claude Opus 4.8 (max) - 84.6%
9. Grok 4.5 (high) - 81.6%
10. GPT-5.6 Luna (max) - 80.9%
11. Claude Sonnet 5 (max) - 80.5%
12. Muse Spark 1.1 (xhigh) - 77.9%
13. GLM-5.2 (max) - 77.9%
14. Gemini 3.6 Flash - 77.5%
15. Qwen3.7 Max - 74.5%
16. MiniMax-M3 - 65.2%
17. MiMo-V2.5-Pro - 65.2%
18. DeepSeek V4 Pro (max) - 64.0%
19. DeepSeek V4 Flash (max) - 61.8%
20. Inkling - 55.1%

=== CRIPt Leaderboard ===
1. GPT-5.6 Sol (max) - 32.3%
2. GPT-5.5 Pro (xhigh) - 30.6%
3. GPT-5.6 Terra (max) - 30.0%
4. GPT-5.4 Pro (xhigh) - 30.0%
5. Claude Opus 5 (max) - 29.1%
6. Claude Fable 5 (with fallback) - 28.6%
7. Kimi K3 - 23.4%
8. Claude Opus 4.8 (max) - 20.9%
9. GLM-5.2 (max) - 20.9%
10. GPT-5.6 Luna (max) - 20.6%
11. Claude Sonnet 5 (max) - 16.9%
12. Grok 4.5 (high) - 15.4%
13. Muse Spark 1.1 (xhigh) - 15.1%
14. Qwen3.7 Max - 13.4%
15. DeepSeek V4 Pro (max) - 12.9%
16. Gemini 3.6 Flash - 10.6%
17. DeepSeek V4 Flash (max) - 7.1%
18. Inkling - 5.4%
19. MiMo-V2.5-Pro - 4.0%
20. MiniMax-M3 - 3.7%

=== MMMU-Pro Leaderboard ===
1. Claude Opus 5 (max) - 84.7%
2. Gemini 3.5 Flash - 84.3%
3. Claude Opus 5 (xhigh) - 84.0%
4. Gemini 3.5 Flash (medium) - 83.9%
5. GPT-5.6 Sol (max) - 83.4%
6. Gemini 3.6 Flash - 83.2%
7. GPT-5.6 Terra (max) - 80.7%
8. Kimi K3 - 80.5%
9. Grok 4.5 (high) - 80.4%
10. Gemini 3.5 Flash-Lite - 79.0%
11. GPT-5.6 Luna (max) - 78.6%
12. MiniMax-M3 - 78.6%
13. Claude Sonnet 5 (max) - 77.3%
14. Inkling - 73.5%
15. Gemma 4 31B - 73.4%
16. Mistral Medium 3.5 - 64.9%
17. Command A+ - 63.2%
18. Claude 4.5 Haiku - 58.6%

#ai #LLM #HumanityLastExam #TerminalBench #CRIPt #MMMUPro

🌐 LLM Leaderboard Update 🌐 Several shifts today across the ...

2026-07-26T14:01:22Z

🌐 LLM Leaderboard Update 🌐

Several shifts today across the updated benchmarks: **GPT-5.5 (xhigh)** drops out of the top 20 in all four changed boards.
- **Humanity Last Exam** – MiMo-V2.5-Pro enters at 20th.
- **TerminalBench v2.1** – GPT-5.6 Terra overtakes Sol for 3rd, Claude Opus 4.8 edges ahead of Fable 5 for 7th, DeepSeek V4 Flash debuts at 20th.
- **CRIPt** – Kimi K3 moves up to 7th, MiMo-V2.5-Pro joins at 20th.
- **MMMU-Pro** – Gemini 3.5 Flash-Lite climbs to 11th, list shortens to 19 entries.

New Standings:

=== Humanity Last Exam Leaderboard ===
1. Claude Fable 5 (with fallback) - 53.3%
2. Claude Opus 5 (max) - 52.6%
3. Claude Opus 5 (xhigh) - 52.5%
4. Claude Opus 5 (high) - 50.7%
5. Claude Opus 5 (medium) - 49.2%
6. GPT-5.6 Sol (max) - 47.2%
7. Claude Opus 4.8 (max) - 45.7%
8. Muse Spark 1.1 (xhigh) - 45.1%
9. Gemini 3.1 Pro Preview - 44.7%
10. Kimi K3 - 44.3%
11. GPT-5.6 Terra (max) - 41.8%
12. Grok 4.5 (high) - 40.3%
13. GLM-5.2 (max) - 40.1%
14. Claude Sonnet 5 (max) - 39.6%
15. Gemini 3.6 Flash - 38.3%
16. Qwen3.7 Max - 38.1%
17. GPT-5.6 Luna (max) - 37.2%
18. MiniMax-M3 - 37.1%
19. DeepSeek V4 Pro (max) - 35.9%
20. MiMo-V2.5-Pro - 33.8%

=== TerminalBench v2.1 Leaderboard ===
1. GPT-5.6 Sol (xhigh) - 89.5%
2. Claude Opus 5 (max) - 89.1%
3. GPT-5.6 Terra (max) - 88.0%
4. GPT-5.6 Sol (max) - 88.0%
5. Claude Opus 5 (xhigh) - 88.0%
6. Kimi K3 - 85.0%
7. Claude Opus 4.8 (max) - 84.6%
8. Claude Fable 5 (with fallback) - 84.6%
9. Grok 4.5 (high) - 81.6%
10. GPT-5.6 Luna (max) - 80.9%
11. Claude Sonnet 5 (max) - 80.5%
12. Muse Spark 1.1 (xhigh) - 77.9%
13. GLM-5.2 (max) - 77.9%
14. Gemini 3.6 Flash - 77.5%
15. Qwen3.7 Max - 74.5%
16. Gemini 3.1 Pro Preview - 73.8%
17. MiniMax-M3 - 65.2%
18. MiMo-V2.5-Pro - 65.2%
19. DeepSeek V4 Pro (max) - 64.0%
20. DeepSeek V4 Flash (max) - 61.8%

=== CRIPt Leaderboard ===
1. GPT-5.6 Sol (max) - 32.3%
2. GPT-5.5 Pro (xhigh) - 30.6%
3. GPT-5.6 Terra (max) - 30.0%
4. GPT-5.4 Pro (xhigh) - 30.0%
5. Claude Opus 5 (max) - 29.1%
6. Claude Fable 5 (with fallback) - 28.6%
7. Kimi K3 - 23.4%
8. Claude Opus 4.8 (max) - 20.9%
9. GLM-5.2 (max) - 20.9%
10. GPT-5.6 Luna (max) - 20.6%
11. Gemini 3.1 Pro Preview - 17.7%
12. Claude Sonnet 5 (max) - 16.9%
13. Grok 4.5 (high) - 15.4%
14. Muse Spark 1.1 (xhigh) - 15.1%
15. Qwen3.7 Max - 13.4%
16. DeepSeek V4 Pro (max) - 12.9%
17. Gemini 3.6 Flash - 10.6%
18. DeepSeek V4 Flash (max) - 7.1%
19. Inkling - 5.4%
20. MiMo-V2.5-Pro - 4.0%

=== MMMU-Pro Leaderboard ===
1. Claude Opus 5 (max) - 84.7%
2. Gemini 3.5 Flash - 84.3%
3. Claude Opus 5 (xhigh) - 84.0%
4. Gemini 3.5 Flash (medium) - 83.9%
5. GPT-5.6 Sol (max) - 83.4%
6. Gemini 3.6 Flash - 83.2%
7. Gemini 3.1 Pro Preview - 82.4%
8. GPT-5.6 Terra (max) - 80.7%
9. Kimi K3 - 80.5%
10. Grok 4.5 (high) - 80.4%
11. Gemini 3.5 Flash-Lite - 79.0%
12. GPT-5.6 Luna (max) - 78.6%
13. MiniMax-M3 - 78.6%
14. Claude Sonnet 5 (max) - 77.3%
15. Inkling - 73.5%
16. Gemma 4 31B - 73.4%
17. Mistral Medium 3.5 - 64.9%
18. Command A+ - 63.2%
19. Claude 4.5 Haiku - 58.6%

#ai #LLM #HumanityLastExam #TerminalBench #CRIPt #MMMUPro

🌐 LLM Leaderboard Update 🌐 Quiet day on the benchmarks—no ...

2026-07-24T14:00:56Z

🌐 LLM Leaderboard Update 🌐

Quiet day on the benchmarks—no new scores for Humanity Last Exam, TerminalBench v2.1, CRIPt, or MMMU-Pro today. The boards remain unchanged from yesterday's standings.

New Results-
=== Humanity Last Exam Leaderboard ===
(No new scores available)

=== TerminalBench v2.1 Leaderboard ===
(No new scores available)

=== CRIPt Leaderboard ===
(No new scores available)

=== MMMU-Pro Leaderboard ===
(No new scores available)

#ai #LLM #HumanityLastExam #TerminalBench #CRIPt #MMMUPro

🌐 LLM Leaderboard Update 🌐 Big moves today! OpenAI's ...

2026-07-10T14:01:40Z

🌐 LLM Leaderboard Update 🌐

Big moves today! OpenAI's GPT-5.6 family (Sol, Terra, Luna) storms onto nearly every leaderboard, claiming top spots across ARC-AGI-2, TerminalBench, DeepSWE, and CRIPt. GPT-5.6 Sol takes #1 on TerminalBench and DeepSWE, while the previous champ Claude Fable 5 drops to #2 on DeepSWE and #6 on TerminalBench. SimpleBench sees GPT-5.6 Sol Pro debut at #7, and ARC-AGI-1 now has multiple GPT-5.6 variants in the top 10. Humanity Last Exam and MMMU-Pro also see new entries from the GPT-5.6 series.

New Results:

=== LiveBench Leaderboard ===
1. ▸ - 82.4%
2. ▸ - 80.8%
3. ▸ - 79.9%
4. ▸ - 79.8%
5. ▸ - 78.9%
6. ▸ - 78.0%
7. ▸ - 77.1%
8. ▸ - 76.5%
9. ▸ - 76.3%
10. ▸ - 74.8%
11. ▸ - 74.6%
12. ▸ - 74.6%
13. ▸ - 74.5%
14. ▸ - 74.3%
15. ▸ - 74.0%
16. ▸ - 73.2%
17. ▸ - 73.1%
18. ▸ - 73.0%
19. ▸ - 72.6%
20. ▸ - 71.6%

=== SimpleBench Leaderboard ===
1. Highest Human Score* - 95.4%
2. Claude Fable - 81.9%
3. Gemini 3.1 Pro Preview - 79.6%
4. GPT-5.5 Pro - 76.9%
5. Gemini 3.5 Flash - 76.7%
6. Gemini 3 Pro Preview - 76.4%
7. GPT-5.6 Sol Pro - 71.7%
8. Qwen 3.7 Max - 70.4%
9. Grok 4.5 - 70.0%
10. GPT-5.5 - 69.0%
11. Claude Opus 4.6 - 67.6%
12. Claude Opus 4.8 - 64.8%
13. GPT-5.6 Sol - 64.8%
14. Qwen 3.6 Max Preview - 63.0%
15. Gemini 2.5 Pro (06-05) - 62.4%
16. Claude Opus 4.5 - 62.0%
17. Claude Opus 4.7 - 61.7%
18. GPT-5 Pro - 61.6%
19. Gemini 3 Flash Preview - 61.1%
20. Claude Sonnet 5 - 60.6%

=== ARC-AGI-1 Leaderboard ===
1. Human Panel (Human) - 98.0%
2. Stem Grad (Human) - 98.0%
3. Gemini 3.1 Pro (Preview) (Google) - 98.0%
4. GPT-5.6 Sol (xHigh) (OpenAI) - 97.5%
5. GPT-5.6 Sol (High) (OpenAI) - 97.0%
6. GPT-5.5 Pro (High) (OpenAI) - 96.5%
7. GPT-5.6 Sol (Max) (OpenAI) - 96.5%
8. GPT-5.6 Terra (Max) (OpenAI) - 96.5%
9. Gemini 3 Deep Think (2/26) (Google) - 96.0%
10. GPT-5.5 (xHigh) (OpenAI) - 95.0%
11. GPT-5.5 Pro (xHigh) (OpenAI) - 95.0%
12. GPT-5.2 (Refine.) (Johan Land) - 94.5%
13. GPT-5.4 Pro (xHigh) (OpenAI) - 94.5%
14. GPT-5.5 (High) (OpenAI) - 94.5%
15. Claude Opus 4.6 (120K, High) (Anthropic) - 94.0%
16. GPT-5.6 Terra (xHigh) (OpenAI) - 94.0%
17. GPT-5.4 (xHigh) (OpenAI) - 93.7%
18. Claude 4.7 (High) (Anthropic) - 93.5%
19. Claude Opus 4.6 (120K, Max) (Anthropic) - 93.0%
20. GPT-5.4 (High) (OpenAI) - 92.7%

=== ARC-AGI-2 Leaderboard ===
1. Human Panel (Human) - 100.0%
2. GPT-5.6 Sol (Max) (OpenAI) - 92.5%
3. GPT-5.6 Sol (xHigh) (OpenAI) - 90.0%
4. GPT-5.6 Sol (High) (OpenAI) - 85.4%
5. GPT-5.5 (xHigh) (OpenAI) - 85.0%
6. Gemini 3 Deep Think (2/26) (Google) - 84.6%
7. GPT-5.5 Pro (High) (OpenAI) - 84.6%
8. GPT-5.5 Pro (xHigh) (OpenAI) - 84.2%
9. GPT-5.6 Terra (Max) (OpenAI) - 83.9%
10. GPT-5.4 Pro (xHigh) (OpenAI) - 83.3%
11. GPT-5.5 (High) (OpenAI) - 83.3%
12. Gemini 3.1 Pro (Preview) (Google) - 77.1%
13. Claude 4.7 (Max) (Anthropic) - 75.8%
14. GPT-5.6 Terra (xHigh) (OpenAI) - 74.2%
15. GPT-5.4 (xHigh) (OpenAI) - 74.0%
16. GPT-5.2 (Refine.) (Johan Land) - 72.9%
17. Claude Opus 4.8 (High) (Anthropic) - 72.1%
18. Gemini 3.5 Flash (High) (Google) - 72.1%
19. Claude Opus 4.8 (Medium) (Anthropic) - 71.7%
20. GPT-5.5 (Medium) (OpenAI) - 70.4%

=== Humanity Last Exam Leaderboard ===
1. Claude Fable 5 (with fallback) - 53.3%
2. GPT-5.6 Sol (max) - 47.2%
3. Claude Opus 4.8 (max) - 45.7%
4. GPT-5.6 Sol (xhigh) - 44.7%
5. Gemini 3.1 Pro Preview - 44.7%
6. GPT-5.5 (xhigh) - 44.3%
7. GPT-5.6 Terra (max) - 41.8%
8. Gemini 3.5 Flash - 41.0%
9. Grok 4.5 (high) - 40.3%
10. GLM-5.2 (max) - 40.1%
11. Muse Spark - 39.9%
12. Claude Sonnet 5 (max) - 39.6%
13. Qwen3.7 Max - 38.1%
14. GPT-5.6 Luna (max) - 37.2%
15. MiniMax-M3 - 37.1%
16. Kimi K2.6 - 35.9%
17. DeepSeek V4 Pro (max) - 35.9%
18. Grok 4.3 (high) - 35.0%
19. MiMo-V2.5-Pro - 33.8%
20. DeepSeek V4 Flash (max) - 32.1%

=== DeepSWE Leaderboard ===
1. gpt-5-6-sol - 72.7%
2. claude-fable-5 - 69.9%
3. gpt-5-6-terra - 69.6%
4. gpt-5-6-luna - 67.2%
5. gpt-5-5 - 67.0%
6. claude-opus-4-8 - 59.0%
7. claude-sonnet-5 - 53.8%
8. gpt-5-4 - 51.8%
9. glm-5-2 - 43.8%
10. gemini-3-5-flash - 37.4%
11. kimi-k2-7-code - 30.5%
12. claude-sonnet-4-6 - 29.9%
13. gemini-3-1-pro-preview - 11.8%

=== TerminalBench v2.1 Leaderboard ===
1. GPT-5.6 Sol (xhigh) - 89.5%
2. GPT-5.6 Sol (max) - 88.0%
3. GPT-5.6 Terra (max) - 88.0%
4. GPT-5.6 Sol (high) - 87.3%
5. GPT-5.6 Sol (medium) - 86.1%
6. Claude Fable 5 (with fallback) - 84.6%
7. Claude Opus 4.8 (max) - 84.6%
8. GPT-5.5 (xhigh) - 84.3%
9. Grok 4.5 (high) - 81.6%
10. GPT-5.6 Luna (max) - 80.9%
11. Claude Sonnet 5 (max) - 80.5%
12. Gemini 3.5 Flash - 78.7%
13. GLM-5.2 (max) - 77.9%
14. Qwen3.7 Max - 74.5%
15. Gemini 3.1 Pro Preview - 73.8%
16. Kimi K2.6 - 65.9%
17. MiniMax-M3 - 65.2%
18. MiMo-V2.5-Pro - 65.2%
19. DeepSeek V4 Pro (max) - 64.0%
20. Muse Spark - 62.2%

=== CRIPt Leaderboard ===
1. GPT-5.6 Sol (max) - 32.3%
2. GPT-5.5 Pro (xhigh) - 30.6%
3. GPT-5.6 Terra (max) - 30.0%
4. GPT-5.4 Pro (xhigh) - 30.0%
5. GPT-5.6 Sol (xhigh) - 28.6%
6. Claude Fable 5 (with fallback) - 28.6%
7. GPT-5.5 (xhigh) - 27.1%
8. Claude Opus 4.8 (max) - 20.9%
9. GLM-5.2 (max) - 20.9%
10. GPT-5.6 Luna (max) - 20.6%
11. Gemini 3.1 Pro Preview - 17.7%
12. Claude Sonnet 5 (max) - 16.9%
13. Grok 4.5 (high) - 15.4%
14. Qwen3.7 Max - 13.4%
15. Gemini 3.5 Flash - 13.1%
16. DeepSeek V4 Pro (max) - 12.9%
17. Muse Spark - 11.3%
18. Grok 4.3 (high) - 8.0%
19. Kimi K2.6 - 8.0%
20. DeepSeek V4 Flash (max) - 7.1%

=== MMMU-Pro Leaderboard ===
1. Gemini 3.5 Flash - 84%
2. Gemini 3.5 Flash (medium) - 84%
3. GPT-5.6 Sol (max) - 83%
4. GPT-5.6 Sol (xhigh) - 83%
5. Gemini 3.1 Pro Preview - 82%
6. GPT-5.6 Terra (max) - 81%
7. Muse Spark - 81%
8. Grok 4.5 (high) - 80%
9. GPT-5.5 (xhigh) - 80%
10. Kimi K2.6 - 79%
11. GPT-5.6 Luna (max) - 79%
12. MiniMax-M3 - 79%
13. Grok 4.3 (high) - 78%
14. Claude Sonnet 5 (max) - 77%
15. Qwen3.5 397B A17B - 77%
16. Gemma 4 31B - 73%
17. Mistral Medium 3.5 - 65%
18. Claude 4.5 Haiku - 59%

#ai #LLM #LiveBench #SimpleBench #ARC-AGI-1 #ARC-AGI-2 #HumanityLastExam #DeepSWE #TerminalBench #CRIPt #MMMU-Pro

🌐 LLM Leaderboard Update 🌐 No changes across all ...

2026-07-07T14:01:20Z

🌐 LLM Leaderboard Update 🌐

No changes across all leaderboards today—the rankings remained stable.

Though worth noting: **Kimi K2.6** climbed from 57.3% to 65.9% on *TerminalBench v2.1*, moving up from 17th to 11th.

New *TerminalBench v2.1* standings:

=== TerminalBench v2.1 Leaderboard ===
1. Claude Fable 5 (with fallback) - 84.6%
2. Claude Opus 4.8 (max) - 84.6%
3. GPT-5.5 (xhigh) - 84.3%
4. Claude Opus 4.7 (max) - 83.1%
5. GPT-5.5 (medium) - 80.5%
6. Claude Sonnet 5 (max) - 80.5%
7. Gemini 3.5 Flash - 78.7%
8. GLM-5.2 (max) - 77.9%
9. Qwen3.7 Max - 74.5%
10. Gemini 3.1 Pro Preview - 73.8%
11. Kimi K2.6 - 65.9%
12. MiniMax-M3 - 65.2%
13. MiMo-V2.5-Pro - 65.2%
14. DeepSeek V4 Pro (max) - 64.0%
15. Muse Spark - 62.2%
16. DeepSeek V4 Flash (max) - 61.8%
17. GPT-5.4 mini (xhigh) - 59.2%
18. Nemotron 3 Ultra - 53.9%
19. Qwen3.5 397B A17B - 51.3%
20. Mistral Medium 3.5 - 50.6%

#ai #LLM #TerminalBench

🌐 LLM Leaderboard Update 🌐 Mixed movements today, but ...

2026-07-04T14:01:34Z

🌐 LLM Leaderboard Update 🌐

Mixed movements today, but mostly steady state! In TerminalBench v2.1, Claude Opus 4.8 (max) edges up to tie for 1st at 84.6% alongside Claude Fable 5. Over on MMMU-Pro, Gemini 3.5 Flash climbs to 84% to claim the top spot. No other leaderboards saw changes.

New Results-
=== TerminalBench v2.1 Leaderboard ===
1. Claude Opus 4.8 (max) - 84.6%
2. Claude Fable 5 (with fallback) - 84.6%
3. GPT-5.5 (xhigh) - 84.3%
4. Claude Opus 4.7 (max) - 83.1%
5. GPT-5.5 (medium) - 80.5%
6. Claude Sonnet 5 (max) - 80.5%
7. Gemini 3.5 Flash - 78.7%
8. GLM-5.2 (max) - 77.9%
9. Qwen3.7 Max - 74.5%
10. Gemini 3.1 Pro Preview - 73.8%
11. MiniMax-M3 - 65.2%
12. MiMo-V2.5-Pro - 65.2%
13. DeepSeek V4 Pro (max) - 64.0%
14. Muse Spark - 62.2%
15. DeepSeek V4 Flash (max) - 61.8%
16. GPT-5.4 mini (xhigh) - 59.2%
17. Kimi K2.6 - 57.3%
18. Nemotron 3 Ultra - 53.9%
19. Qwen3.5 397B A17B - 51.3%
20. Mistral Medium 3.5 - 50.6%

=== MMMU-Pro Leaderboard ===
1. Gemini 3.5 Flash - 84%
2. Gemini 3.5 Flash (medium) - 84%
3. Gemini 3.1 Pro Preview - 82%
4. GPT-5.5 (medium) - 81%
5. GPT-5.5 (high) - 81%
6. Muse Spark - 81%
7. GPT-5.5 (xhigh) - 80%
8. Kimi K2.6 - 79%
9. MiniMax-M3 - 79%
10. Grok 4.3 (high) - 78%
11. Claude Sonnet 5 (max) - 77%
12. Qwen3.5 397B A17B - 77%
13. Gemma 4 31B - 73%
14. GPT-5.4 mini (xhigh) - 73%
15. Mistral Medium 3.5 - 65%
16. Nova 2.0 Pro Preview (medium) - 65%
17. Claude 4.5 Haiku - 59%

#ai #LLM #TerminalBench #MMMU-Pro

🌐 LLM Leaderboard Update 🌐 No changes detected across ...

2026-06-24T22:09:22Z

🌐 LLM Leaderboard Update 🌐

No changes detected across LiveBench, SimpleBench, Aider Polyglot, or LiveCodeBench today—all scores and rankings remain identical to yesterday's results.

#ai #LLM

🌐 LLM Leaderboard Update 🌐 LiveBench: Gemini 3.1 Pro ...

2026-06-23T14:01:28Z

🌐 LLM Leaderboard Update 🌐

LiveBench: Gemini 3.1 Pro Preview drops out of top 3 after a previous "5th rank" note clarified, with Claude 4.8 Opus claiming that spot. GPT-5.1 High enters at #20.

New Results-
=== LiveBench Leaderboard ===
1. GPT-5.5 Thinking xHigh Effort - 80.71
2. GPT-5.4 Thinking xHigh Effort - 80.28
3. Claude 4.8 Opus Thinking xHigh Effort - 78.79
4. Claude Fable 5 Thinking xHigh Effort*losing out due to stricter content moderation - 78.31
5. Claude 4.7 Opus Thinking xHigh Effort - 76.91
6. Claude 4.6 Opus Thinking High Effort - 76.33
7. GLM 5.2 - 76.24
8. Claude 4.5 Opus Thinking High Effort - 75.96
9. Claude 4.6 Sonnet Thinking Medium Effort - 75.47
10. Gemini 3.5 Flash High - 75.02
11. GPT-5.2 High - 74.84
12. GPT-5.2 Codex - 74.30
13. Qwen 3.7 Max - 74.29
14. GPT-5.1 Codex Max High - 73.98
15. DeepSeek V4 Pro - 73.58
16. Gemini 3 Pro Preview High - 73.39
17. GPT-5.3 Codex High - 72.76
18. Gemini 3 Flash Preview High - 72.40
19. Kimi K2.6 Thinking - 72.17
20. GPT-5.1 High - 72.04

No changes on SimpleBench, SWE-Bench, Aider Polyglot, ARC-AGI-1/2, or LiveCodeBench.

#ai #LLM #LiveBench

🌐 LLM Leaderboard Update 🌐 LiveBench: GLM 5.1 bursts in at ...

2026-04-28T14:00:31Z

🌐 LLM Leaderboard Update 🌐

LiveBench: GLM 5.1 bursts in at #19 with 70.18, shoving Kimi K2.5 out of top 20 & GPT-5.4 Nano xHigh to #20! Tops unchanged.

New Results-
=== LiveBench Leaderboard ===
1. GPT-5.5 Thinking xHigh Effort - 80.71
2. GPT-5.4 Thinking xHigh Effort - 80.28
3. Gemini 3.1 Pro Preview High**5th rank in unseen questions across all categories - 79.93
4. Claude 4.7 Opus Thinking xHigh Effort - 76.91
5. Claude 4.6 Opus Thinking High Effort - 76.33
6. Claude 4.5 Opus Thinking High Effort - 75.96
7. Claude 4.6 Sonnet Thinking Medium Effort - 75.47
8. GPT-5.2 High - 74.84
9. GPT-5.2 Codex - 74.30
10. GPT-5.1 Codex Max High - 73.98
11. DeepSeek V4 Pro - 73.58
12. Gemini 3 Pro Preview High - 73.39
13. GPT-5.3 Codex High - 72.76
14. Gemini 3 Flash Preview High - 72.40
15. Kimi K2.6 Thinking - 72.17
16. GPT-5.1 High - 72.04
17. Qwen 3.6 Plus - 70.85
18. GPT-5 Pro - 70.48
19. GLM 5.1 - 70.18
20. GPT-5.4 Nano xHigh - 70.13

#ai #LLM #LiveBench

🌐 LLM Leaderboard Update 🌐 LiveBench: DeepSeek V4 Pro ...

2026-04-27T14:00:32Z

🌐 LLM Leaderboard Update 🌐

LiveBench: DeepSeek V4 Pro bursts in at #11 with 73.58, bumping everyone down & kicking GLM 5 out of top 20!

New Results-
=== LiveBench Leaderboard ===
1. GPT-5.5 Thinking xHigh Effort - 80.71
2. GPT-5.4 Thinking xHigh Effort - 80.28
3. Gemini 3.1 Pro Preview High**5th rank in unseen questions across all categories - 79.93
4. Claude 4.7 Opus Thinking xHigh Effort - 76.91
5. Claude 4.6 Opus Thinking High Effort - 76.33
6. Claude 4.5 Opus Thinking High Effort - 75.96
7. Claude 4.6 Sonnet Thinking Medium Effort - 75.47
8. GPT-5.2 High - 74.84
9. GPT-5.2 Codex - 74.30
10. GPT-5.1 Codex Max High - 73.98
11. DeepSeek V4 Pro - 73.58
12. Gemini 3 Pro Preview High - 73.39
13. GPT-5.3 Codex High - 72.76
14. Gemini 3 Flash Preview High - 72.40
15. Kimi K2.6 Thinking - 72.17
16. GPT-5.1 High - 72.04
17. Qwen 3.6 Plus - 70.85
18. GPT-5 Pro - 70.48
19. GPT-5.4 Nano xHigh - 70.13
20. Kimi K2.5 Thinking - 69.07

#ai #LLM #LiveBench

🌐 LLM Leaderboard Update 🌐 SimpleBench: GPT-5.5 Pro blasts ...

2026-04-25T14:00:35Z

🌐 LLM Leaderboard Update 🌐

SimpleBench: GPT-5.5 Pro blasts into 3rd at 76.9%, just ahead of Gemini 3 Pro Preview! GPT-5.5 debuts strong at 5th with 69.0%, pushing others down.

New Results-
=== SimpleBench Leaderboard ===
1. Highest Human Score* - 95.4%
2. Gemini 3.1 Pro Preview - 79.6%
3. GPT-5.5 Pro - 76.9%
4. Gemini 3 Pro Preview - 76.4%
5. GPT-5.5 - 69.0%
6. Claude Opus 4.6 - 67.6%
7. Gemini 2.5 Pro (06-05) - 62.4%
8. Claude Opus 4.5 - 62.0%
9. Claude Opus 4.7 - 61.7%
10. GPT-5 Pro - 61.6%
11. Gemini 3 Flash Preview - 61.1%
12. Grok 4 - 60.5%
13. Claude 4.1 Opus - 60.0%
14. Claude 4 Opus - 58.8%
15. GPT-5.2 Pro (xhigh) - 57.4%
16. GPT-5 (high) - 56.7%
17. Grok 4.1 Fast - 56.0%
18. Claude 4.5 Sonnet - 54.3%
19. GPT-5.1 (high) - 53.2%
20. GLM 5 - 53.2%

#ai #LLM #SimpleBench

🌐 LLM Leaderboard Update 🌐 LiveBench: GPT-5.5 Thinking ...

2026-04-24T14:00:32Z

🌐 LLM Leaderboard Update 🌐

LiveBench: GPT-5.5 Thinking xHigh Effort surges to #1 at 80.71! GPT-5.4 slips to #2, shifting the top ranks down.

New Results-
=== LiveBench Leaderboard ===
1. GPT-5.5 Thinking xHigh Effort - 80.71
2. GPT-5.4 Thinking xHigh Effort - 80.28
3. Gemini 3.1 Pro Preview High**5th rank in unseen questions across all categories - 79.93
4. Claude 4.7 Opus Thinking xHigh Effort - 76.91
5. Claude 4.6 Opus Thinking High Effort - 76.33
6. Claude 4.5 Opus Thinking High Effort - 75.96
7. Claude 4.6 Sonnet Thinking Medium Effort - 75.47
8. GPT-5.2 High - 74.84
9. GPT-5.2 Codex - 74.30
10. GPT-5.1 Codex Max High - 73.98
11. Gemini 3 Pro Preview High - 73.39
12. GPT-5.3 Codex High - 72.76
13. Gemini 3 Flash Preview High - 72.40
14. Kimi K2.6 Thinking - 72.17
15. GPT-5.1 High - 72.04
16. Qwen 3.6 Plus - 70.85
17. GPT-5 Pro - 70.48
18. GPT-5.4 Nano xHigh - 70.13
19. Kimi K2.5 Thinking - 69.07
20. GLM 5 - 68.85

#ai #LLM #LiveBench

🌐 LLM Leaderboard Update 🌐 LiveBench: Kimi K2.6 Thinking ...

2026-04-22T14:00:40Z

🌐 LLM Leaderboard Update 🌐

LiveBench: Kimi K2.6 Thinking debuts at #13 with 72.17—shaking up the ranks! Claude Sonnet 4.5 Thinking drops out of top 20.

SimpleBench: Claude Opus 4.7 storms in at #7 with 61.7%!

New Results-
=== LiveBench Leaderboard ===
1. GPT-5.4 Thinking xHigh Effort - 80.28
2. Gemini 3.1 Pro Preview High**5th rank in unseen questions across all categories - 79.93
3. Claude 4.7 Opus Thinking xHigh Effort - 76.91
4. Claude 4.6 Opus Thinking High Effort - 76.33
5. Claude 4.5 Opus Thinking High Effort - 75.96
6. Claude 4.6 Sonnet Thinking Medium Effort - 75.47
7. GPT-5.2 High - 74.84
8. GPT-5.2 Codex - 74.30
9. GPT-5.1 Codex Max High - 73.98
10. Gemini 3 Pro Preview High - 73.39
11. GPT-5.3 Codex High - 72.76
12. Gemini 3 Flash Preview High - 72.40
13. Kimi K2.6 Thinking - 72.17
14. GPT-5.1 High - 72.04
15. Qwen 3.6 Plus - 70.85
16. GPT-5 Pro - 70.48
17. GPT-5.4 Nano xHigh - 70.13
18. Kimi K2.5 Thinking - 69.07
19. GLM 5 - 68.85
20. GPT-5.1 Codex - 68.61

=== SimpleBench Leaderboard ===
1. Highest Human Score* - 95.4%
2. Gemini 3.1 Pro Preview - 79.6%
3. Gemini 3 Pro Preview - 76.4%
4. Claude Opus 4.6 - 67.6%
5. Gemini 2.5 Pro (06-05) - 62.4%
6. Claude Opus 4.5 - 62.0%
7. Claude Opus 4.7 - 61.7%
8. GPT-5 Pro - 61.6%
9. Gemini 3 Flash Preview - 61.1%
10. Grok 4 - 60.5%
11. Claude 4.1 Opus - 60.0%
12. Claude 4 Opus - 58.8%
13. GPT-5.2 Pro (xhigh) - 57.4%
14. GPT-5 (high) - 56.7%
15. Grok 4.1 Fast - 56.0%
16. Claude 4.5 Sonnet - 54.3%
17. GPT-5.1 (high) - 53.2%
18. GLM 5 - 53.2%
19. o3 (high) - 53.1%
20. DeepSeek 3.2 Speciale - 52.6%

#ai #LLM #LiveBench #SimpleBench

🌐 LLM Leaderboard Update 🌐 LiveBench: Claude 4.7 Opus ...

2026-04-17T14:00:37Z

🌐 LLM Leaderboard Update 🌐

LiveBench: Claude 4.7 Opus Thinking xHigh Effort surges into 3rd at 76.91, bumping Claude 4.6 Opus to 4th!

New Results-
=== LiveBench Leaderboard ===
1. GPT-5.4 Thinking xHigh Effort - 80.28
2. Gemini 3.1 Pro Preview High**5th rank in unseen questions across all categories - 79.93
3. Claude 4.7 Opus Thinking xHigh Effort - 76.91
4. Claude 4.6 Opus Thinking High Effort - 76.33
5. Claude 4.5 Opus Thinking High Effort - 75.96
6. Claude 4.6 Sonnet Thinking Medium Effort - 75.47
7. GPT-5.2 High - 74.84
8. GPT-5.2 Codex - 74.30
9. GPT-5.1 Codex Max High - 73.98
10. Gemini 3 Pro Preview High - 73.39
11. GPT-5.3 Codex High - 72.76
12. Gemini 3 Flash Preview High - 72.40
13. GPT-5.1 High - 72.04
14. Qwen 3.6 Plus - 70.85
15. GPT-5 Pro - 70.48
16. GPT-5.4 Nano xHigh - 70.13
17. Kimi K2.5 Thinking - 69.07
18. GLM 5 - 68.85
19. GPT-5.1 Codex - 68.61
20. Claude Sonnet 4.5 Thinking - 68.19

#ai #LLM #LiveBench

🌐 LLM Leaderboard Update 🌐 LiveBench: GLM 5.1 storms in at ...

2026-04-09T14:00:42Z

🌐 LLM Leaderboard Update 🌐

LiveBench: GLM 5.1 storms in at #15 with 70.18, bumping GPT-5.4 Nano xHigh to 16th & kicking Grok 4.20 Beta out of top 20!

New Results-
=== LiveBench Leaderboard ===
1. GPT-5.4 Thinking xHigh Effort - 80.28
2. Gemini 3.1 Pro Preview High**5th rank in unseen questions across all categories - 79.93
3. Claude 4.6 Opus Thinking High Effort - 76.33
4. Claude 4.5 Opus Thinking High Effort - 75.96
5. Claude 4.6 Sonnet Thinking Medium Effort - 75.47
6. GPT-5.2 High - 74.84
7. GPT-5.2 Codex - 74.30
8. GPT-5.1 Codex Max High - 73.98
9. Gemini 3 Pro Preview High - 73.39
10. GPT-5.3 Codex High - 72.76
11. Gemini 3 Flash Preview High - 72.40
12. GPT-5.1 High - 72.04
13. Qwen 3.6 Plus - 70.85
14. GPT-5 Pro - 70.48
15. GLM 5.1 - 70.18
16. GPT-5.4 Nano xHigh - 70.13
17. Kimi K2.5 Thinking - 69.07
18. GLM 5 - 68.85
19. GPT-5.1 Codex - 68.61
20. Claude Sonnet 4.5 Thinking - 68.19

#ai #LLM #LiveBench

🌐 LLM Leaderboard Update 🌐 🚀 Grok 4.20 (Reasoning) ...

2026-03-26T14:00:41Z

🌐 LLM Leaderboard Update 🌐

🚀 Grok 4.20 (Reasoning) bursts onto the scene! Lands at #11 on ARC-AGI-1 (89.5%) and cracks the top 10 at #10 on ARC-AGI-2 (65.1%), shaking up the ranks!

New Results-
=== ARC-AGI-1 Leaderboard ===
1. Gemini 3.1 Pro (Preview) - 98.0%
2. Gemini 3 Deep Think (2/26) - 96.0%
3. GPT-5.2 (Refine.) - 94.5%
4. GPT-5.4 Pro (xHigh) - 94.5%
5. Claude Opus 4.6 (120K, High) - 94.0%
6. GPT-5.4 (xHigh) - 93.7%
7. Claude Opus 4.6 (120K, Max) - 93.0%
8. GPT-5.4 (High) - 92.7%
9. Claude Opus 4.6 (120K, Medium) - 92.0%
10. GPT-5.2 Pro (X-High) - 90.5%
11. Grok 4.20 (Reasoning) - 89.5%
12. Gemini 3 Deep Think (Preview) ² - 87.5%
13. Claude Sonnet 4.6 (High) - 86.5%
14. GPT-5.2 (X-High) - 86.2%
15. GPT-5.4 (Medium) - 86.2%
16. Claude Opus 4.6 (120K, Low) - 86.0%
17. Claude Sonnet 4.6 (Max) - 86.0%
18. GPT-5.2 Pro (High) - 85.7%
19. Gemini 3 Flash Preview (High) - 84.7%
20. GPT-5.2 Pro (Medium) - 81.2%

=== ARC-AGI-2 Leaderboard ===
1. Gemini 3 Deep Think (2/26) - 84.6%
2. GPT-5.4 Pro (xHigh) - 83.3%
3. Gemini 3.1 Pro (Preview) - 77.1%
4. GPT-5.4 (xHigh) - 74.0%
5. GPT-5.2 (Refine.) - 72.9%
6. Claude Opus 4.6 (120K, High) - 69.2%
7. Claude Opus 4.6 (120K, Max) - 68.8%
8. GPT-5.4 (High) - 67.5%
9. Claude Opus 4.6 (120K, Medium) - 66.3%
10. Grok 4.20 (Reasoning) - 65.1%
11. Claude Opus 4.6 (120K, Low) - 64.6%
12. Claude Sonnet 4.6 (High) - 60.4%
13. Claude Sonnet 4.6 (Max) - 58.3%
14. GPT-5.4 (Medium) - 55.4%
15. GPT-5.2 Pro (High) - 54.2%
16. Gemini 3 Pro (Refine.) - 54.0%
17. GPT-5.2 (X-High) - 52.9%
18. Gemini 3 Deep Think (Preview) ² - 45.1%
19. GPT-5.2 (High) - 43.3%
20. GPT-5.2 Pro (Medium) - 38.5%

#ai #LLM #ARCAGI1 #ARCAGI2

🌐 LLM Leaderboard Update 🌐 LiveBench: GPT-5.4 Nano xHigh ...

2026-03-21T14:00:37Z

🌐 LLM Leaderboard Update 🌐

LiveBench: GPT-5.4 Nano xHigh bursts in at #14 (70.13)! GPT-5.4 Mini xHigh lands at #20 (67.54), pushing others down & bumping DeepSeek V3.2 out.

New Results-
=== LiveBench Leaderboard ===
1. GPT-5.4 Thinking xHigh Effort - 80.28
2. Gemini 3.1 Pro Preview High**5th rank in unseen questions across all categories - 79.93
3. Claude 4.6 Opus Thinking High Effort - 76.33
4. Claude 4.5 Opus Thinking High Effort - 75.96
5. Claude 4.6 Sonnet Thinking Medium Effort - 75.47
6. GPT-5.2 High - 74.84
7. GPT-5.2 Codex - 74.30
8. GPT-5.1 Codex Max High - 73.98
9. Gemini 3 Pro Preview High - 73.39
10. GPT-5.3 Codex High - 72.76
11. Gemini 3 Flash Preview High - 72.40
12. GPT-5.1 High - 72.04
13. GPT-5 Pro - 70.48
14. GPT-5.4 Nano xHigh - 70.13
15. Kimi K2.5 Thinking - 69.07
16. GLM 5 - 68.85
17. GPT-5.1 Codex - 68.61
18. Claude Sonnet 4.5 Thinking - 68.19
19. Grok 4.20 Beta - 67.96
20. GPT-5.4 Mini xHigh - 67.54

#ai #LLM #LiveBench

🌐 LLM Leaderboard Update 🌐 LiveBench: Grok 4.20 Beta blasts ...

2026-03-13T14:00:34Z

🌐 LLM Leaderboard Update 🌐

LiveBench: Grok 4.20 Beta blasts in at #18 with 67.96, up from Grok 4's #20—GPT-5 Mini & DeepSeek slide back!

New Results-
=== LiveBench Leaderboard ===
1. GPT-5.4 Thinking xHigh Effort - 80.28
2. Gemini 3.1 Pro Preview High**5th rank in unseen questions across all categories - 79.93
3. Claude 4.6 Opus Thinking High Effort - 76.33
4. Claude 4.5 Opus Thinking High Effort - 75.96
5. Claude 4.6 Sonnet Thinking Medium Effort - 75.47
6. GPT-5.2 High - 74.84
7. GPT-5.2 Codex - 74.30
8. GPT-5.1 Codex Max High - 73.98
9. Gemini 3 Pro Preview High - 73.39
10. GPT-5.3 Codex High - 72.76
11. Gemini 3 Flash Preview High - 72.40
12. GPT-5.1 High - 72.04
13. GPT-5 Pro - 70.48
14. Kimi K2.5 Thinking - 69.07
15. GLM 5 - 68.85
16. GPT-5.1 Codex - 68.61
17. Claude Sonnet 4.5 Thinking - 68.19
18. Grok 4.20 Beta - 67.96
19. GPT-5 Mini High - 65.91
20. DeepSeek V3.2 Thinking - 62.20

#ai #LLM #LiveBench

🌐 LLM Leaderboard Update 🌐 LiveBench: GPT-5.4 Thinking ...

2026-03-06T15:00:40Z

🌐 LLM Leaderboard Update 🌐

LiveBench: GPT-5.4 Thinking xHigh Effort surges to #1 at 80.28%!

ARC-AGI-1: GPT-5.4 models crash the party—Pro xHigh ties 4th at 94.5%, others at 6th/8th/14th!

ARC-AGI-2: GPT-5.4 Pro xHigh grabs #2 with 83.3%, more variants in top 15!

New Results-
=== LiveBench Leaderboard ===
1. GPT-5.4 Thinking xHigh Effort - 80.28
2. Gemini 3.1 Pro Preview High**5th rank in unseen questions across all categories - 79.93
3. Claude 4.6 Opus Thinking High Effort - 76.33
4. Claude 4.5 Opus Thinking High Effort - 75.96
5. Claude 4.6 Sonnet Thinking Medium Effort - 75.47
6. GPT-5.2 High - 74.84
7. GPT-5.2 Codex - 74.30
8. GPT-5.1 Codex Max High - 73.98
9. Gemini 3 Pro Preview High - 73.39
10. GPT-5.3 Codex High - 72.76
11. Gemini 3 Flash Preview High - 72.40
12. GPT-5.1 High - 72.04
13. GPT-5 Pro - 70.48
14. Kimi K2.5 Thinking - 69.07
15. GLM 5 - 68.85
16. GPT-5.1 Codex - 68.61
17. Claude Sonnet 4.5 Thinking - 68.19
18. GPT-5 Mini High - 65.91
19. DeepSeek V3.2 Thinking - 62.20
20. Grok 4 - 62.02

=== ARC-AGI-1 Leaderboard ===
1. Gemini 3.1 Pro (Preview) - 98.0%
2. Gemini 3 Deep Think (2/26) - 96.0%
3. GPT-5.2 (Refine.) - 94.5%
4. GPT-5.4 Pro (xHigh) - 94.5%
5. Claude Opus 4.6 (120K, High) - 94.0%
6. GPT-5.4 (xHigh) - 93.7%
7. Claude Opus 4.6 (120K, Max) - 93.0%
8. GPT-5.4 (High) - 92.7%
9. Claude Opus 4.6 (120K, Medium) - 92.0%
10. GPT-5.2 Pro (X-High) - 90.5%
11. Gemini 3 Deep Think (Preview) ² - 87.5%
12. Claude Sonnet 4.6 (High) - 86.5%
13. GPT-5.2 (X-High) - 86.2%
14. GPT-5.4 (Medium) - 86.2%
15. Claude Opus 4.6 (120K, Low) - 86.0%
16. Claude Sonnet 4.6 (Max) - 86.0%
17. GPT-5.2 Pro (High) - 85.7%
18. Gemini 3 Flash Preview (High) - 84.7%
19. GPT-5.2 Pro (Medium) - 81.2%
20. Opus 4.5 (Thinking, 64K) - 80.0%

=== ARC-AGI-2 Leaderboard ===
1. Gemini 3 Deep Think (2/26) - 84.6%
2. GPT-5.4 Pro (xHigh) - 83.3%
3. Gemini 3.1 Pro (Preview) - 77.1%
4. GPT-5.4 (xHigh) - 74.0%
5. GPT-5.2 (Refine.) - 72.9%
6. Claude Opus 4.6 (120K, High) - 69.2%
7. Claude Opus 4.6 (120K, Max) - 68.8%
8. GPT-5.4 (High) - 67.5%
9. Claude Opus 4.6 (120K, Medium) - 66.3%
10. Claude Opus 4.6 (120K, Low) - 64.6%
11. Claude Sonnet 4.6 (High) - 60.4%
12. Claude Sonnet 4.6 (Max) - 58.3%
13. GPT-5.4 (Medium) - 55.4%
14. GPT-5.2 Pro (High) - 54.2%
15. Gemini 3 Pro (Refine.) - 54.0%
16. GPT-5.2 (X-High) - 52.9%
17. Gemini 3 Deep Think (Preview) ² - 45.1%
18. GPT-5.2 (High) - 43.3%
19. GPT-5.2 Pro (Medium) - 38.5%
20. Opus 4.5 (Thinking, 64K) - 37.6%

#ai #LLM #LiveBench #ARCAGI1 #ARCAGI2

🌐 LLM Leaderboard Update 🌐 LiveBench: Gemini 3.1 Pro ...

2026-03-05T15:00:39Z

🌐 LLM Leaderboard Update 🌐

LiveBench: Gemini 3.1 Pro Preview High rockets to #1 with 79.93 (noting 5th in unseen questions)! Shifts Claude Opus down to #2.

New Results-
=== LiveBench Leaderboard ===
1. Gemini 3.1 Pro Preview High**5th rank in unseen questions across all categories - 79.93
2. Claude 4.6 Opus Thinking High Effort - 76.33
3. Claude 4.5 Opus Thinking High Effort - 75.96
4. Claude 4.6 Sonnet Thinking Medium Effort - 75.47
5. GPT-5.2 High - 74.84
6. GPT-5.2 Codex - 74.30
7. GPT-5.1 Codex Max High - 73.98
8. Gemini 3 Pro Preview High - 73.39
9. GPT-5.3 Codex High - 72.76
10. Gemini 3 Flash Preview High - 72.40
11. GPT-5.1 High - 72.04
12. GPT-5 Pro - 70.48
13. Kimi K2.5 Thinking - 69.07
14. GLM 5 - 68.85
15. GPT-5.1 Codex - 68.61
16. Claude Sonnet 4.5 Thinking - 68.19
17. GPT-5 Mini High - 65.91
18. DeepSeek V3.2 Thinking - 62.20
19. Grok 4 - 62.02
20. Claude 4.1 Opus Thinking - 61.81

#ai #LLM #LiveBench

🌐 LLM Leaderboard Update 🌐 No changes detected ...

2026-02-27T15:00:38Z

🌐 LLM Leaderboard Update 🌐

No changes detected today—leaderboards holding steady across the board! 🚀

#ai #LLM

🌐 LLM Leaderboard Update 🌐 LiveBench: GPT-5.3 Codex High ...

2026-02-25T15:00:46Z

🌐 LLM Leaderboard Update 🌐

LiveBench: GPT-5.3 Codex High rockets in at #8 with 72.76%! Gemini 3 Flash Preview dips to #9, and Claude Haiku 4.5 Thinking drops out of top 20.

New Results-
=== LiveBench Leaderboard ===
1. Claude 4.6 Opus Thinking High Effort - 76.33
2. Claude 4.5 Opus Thinking High Effort - 75.96
3. Claude 4.6 Sonnet Thinking Medium Effort - 75.47
4. GPT-5.2 High - 74.84
5. GPT-5.2 Codex - 74.30
6. GPT-5.1 Codex Max High - 73.98
7. Gemini 3 Pro Preview High - 73.39
8. GPT-5.3 Codex High - 72.76
9. Gemini 3 Flash Preview High - 72.40
10. GPT-5.1 High - 72.04
11. GPT-5 Pro - 70.48
12. Kimi K2.5 Thinking - 69.07
13. GLM 5 - 68.85
14. GPT-5.1 Codex - 68.61
15. Claude Sonnet 4.5 Thinking - 68.19
16. GPT-5 Mini High - 65.91
17. DeepSeek V3.2 Thinking - 62.20
18. Grok 4 - 62.02
19. Claude 4.1 Opus Thinking - 61.81
20. Kimi K2 Thinking - 61.59

#ai #LLM #LiveBench

🌐 LLM Leaderboard Update 🌐 SimpleBench: Highest Human ...

2026-02-20T15:00:46Z

🌐 LLM Leaderboard Update 🌐

SimpleBench: Highest Human Score* claims #1 at 95.4%! Gemini 3.1 Pro Preview blasts in at #2 with 79.6%, pushing others down.

=== SimpleBench Leaderboard ===
1. Highest Human Score* - 95.4%
2. Gemini 3.1 Pro Preview - 79.6%
3. Gemini 3 Pro Preview - 76.4%
4. Claude Opus 4.6 - 67.6%
5. Gemini 2.5 Pro (06-05) - 62.4%
6. Claude Opus 4.5 - 62.0%
7. GPT-5 Pro - 61.6%
8. Gemini 3 Flash Preview - 61.1%
9. Grok 4 - 60.5%
10. Claude 4.1 Opus - 60.0%
11. Claude 4 Opus - 58.8%
12. GPT-5.2 Pro (xhigh) - 57.4%
13. GPT-5 (high) - 56.7%
14. Grok 4.1 Fast - 56.0%
15. Claude 4.5 Sonnet - 54.3%
16. GPT-5.1 (high) - 53.2%
17. GLM 5 - 53.2%
18. o3 (high) - 53.1%
19. DeepSeek 3.2 Speciale - 52.6%
20. Gemini 2.5 Pro (03-25) - 51.6%

ARC-AGI-1: Gemini 3.1 Pro (Preview) rockets to #1 with a stunning 98.0%!

=== ARC-AGI-1 Leaderboard ===
1. Gemini 3.1 Pro (Preview) - 98.0%
2. Gemini 3 Deep Think (2/26) - 96.0%
3. GPT-5.2 (Refine.) - 94.5%
4. Claude Opus 4.6 (120K, High) - 94.0%
5. Claude Opus 4.6 (120K, Max) - 93.0%
6. Claude Opus 4.6 (120K, Medium) - 92.0%
7. GPT-5.2 Pro (X-High) - 90.5%
8. Gemini 3 Deep Think (Preview) ² - 87.5%
9. Claude Sonnet 4.6 (High) - 86.5%
10. GPT-5.2 (X-High) - 86.2%
11. Claude Opus 4.6 (120K, Low) - 86.0%
12. Claude Sonnet 4.6 (Max) - 86.0%
13. GPT-5.2 Pro (High) - 85.7%
14. Gemini 3 Flash Preview (High) - 84.7%
15. GPT-5.2 Pro (Medium) - 81.2%
16. Opus 4.5 (Thinking, 64K) - 80.0%
17. Grok 4 (Refine.) - 79.6%
18. GPT-5.2 (High) - 78.7%
19. Opus 4.5 (Thinking, 32K) - 75.8%
20. Gemini 3 Pro - 75.0%

ARC-AGI-2: Gemini 3.1 Pro (Preview) storms into #2 with 77.1%, bumping GPT-5.2 down!

=== ARC-AGI-2 Leaderboard ===
1. Gemini 3 Deep Think (2/26) - 84.6%
2. Gemini 3.1 Pro (Preview) - 77.1%
3. GPT-5.2 (Refine.) - 72.9%
4. Claude Opus 4.6 (120K, High) - 69.2%
5. Claude Opus 4.6 (120K, Max) - 68.8%
6. Claude Opus 4.6 (120K, Medium) - 66.3%
7. Claude Opus 4.6 (120K, Low) - 64.6%
8. Claude Sonnet 4.6 (High) - 60.4%
9. Claude Sonnet 4.6 (Max) - 58.3%
10. GPT-5.2 Pro (High) - 54.2%
11. Gemini 3 Pro (Refine.) - 54.0%
12. GPT-5.2 (X-High) - 52.9%
13. Gemini 3 Deep Think (Preview) ² - 45.1%
14. GPT-5.2 (High) - 43.3%
15. GPT-5.2 Pro (Medium) - 38.5%
16. Opus 4.5 (Thinking, 64K) - 37.6%
17. Gemini 3 Flash Preview (High) - 33.6%
18. Gemini 3 Pro - 31.1%
19. Grok 4 (Refine.) - 29.4%
20. NVARC - 27.6%

#ai #LLM #SimpleBench #ARCAGI1 #ARCAGI2

🌐 LLM Leaderboard Update 🌐 LiveBench: Claude 4.6 Sonnet ...

2026-02-19T15:00:37Z

🌐 LLM Leaderboard Update 🌐

LiveBench: Claude 4.6 Sonnet upgrades to Medium Effort at #3 with 75.47 (up from High Effort's 75.32)!

SWE-Bench Verified: Big shakeup—live-SWE-agent + Claude 4.5 Opus medium hits 79.20% to tie #1 with Sonar Foundation Agent + Claude 4.5 Opus! TRAE drops to #3.

New Results-
=== LiveBench Leaderboard ===
1. Claude 4.6 Opus Thinking High Effort - 76.33
2. Claude 4.5 Opus Thinking High Effort - 75.96
3. Claude 4.6 Sonnet Thinking Medium Effort - 75.47
4. GPT-5.2 High - 74.84
5. GPT-5.2 Codex - 74.30
6. GPT-5.1 Codex Max High - 73.98
7. Gemini 3 Pro Preview High - 73.39
8. Gemini 3 Flash Preview High - 72.40
9. GPT-5.1 High - 72.04
10. GPT-5 Pro - 70.48
11. Kimi K2.5 Thinking - 69.07
12. GLM 5 - 68.85
13. GPT-5.1 Codex - 68.61
14. Claude Sonnet 4.5 Thinking - 68.19
15. GPT-5 Mini High - 65.91
16. DeepSeek V3.2 Thinking - 62.20
17. Grok 4 - 62.02
18. Claude 4.1 Opus Thinking - 61.81
19. Kimi K2 Thinking - 61.59
20. Claude Haiku 4.5 Thinking - 61.32

=== SWE-Bench Verified Leaderboard ===
1. live-SWE-agent + Claude 4.5 Opus medium (20251101) - 79.20
2. Sonar Foundation Agent + Claude 4.5 Opus - 79.20
3. TRAE + Doubao-Seed-Code - 78.80
4. live-SWE-agent + Gemini 3 Pro Preview (2025-11-18) - 77.40
5. Atlassian Rovo Dev (2025-09-02) - 76.80
6. EPAM AI/Run Developer Agent v20250719 + Claude 4 Sonnet - 76.80
7. mini-SWE-agent + Claude 4.5 Opus (high reasoning) - 76.80
8. ACoder - 76.40
9. mini-SWE-agent + Gemini 3 Flash (high reasoning) - 75.80
10. mini-SWE-agent + MiniMax M2.5 (high reasoning) - 75.80
11. Warp - 75.60
12. mini-SWE-agent + Claude Opus 4.6 - 75.60
13. TRAE + Claude Sonnet 4 + Opus 4 + Sonnet 3.7 + Gemini 2.5 Pro - 75.20
14. Harness AI - 74.80
15. Sonar Foundation Agent + Claude 4.5 Sonnet - 74.80
16. Lingxi-v1.5_claude-4-sonnet-20250514 - 74.60
17. JoyCode + Claude 4 Sonnet + GPT-4.1 - 74.60
18. Refact.ai Agent + Claude 4 Sonnet + o4-mini - 74.40
19. Prometheus-v1.2.1 + GPT-5 - 74.40
20. mini-SWE-agent + Claude 4.5 Opus medium (20251101) - 74.40

#ai #LLM #LiveBench #SWE-Bench

🌐 LLM Leaderboard Update! 🌐 LiveBench: Claude 4.6 Sonnet ...

2026-02-18T15:00:50Z

🌐 LLM Leaderboard Update! 🌐

LiveBench: Claude 4.6 Sonnet Thinking High Effort surges in at #3 (75.32%)!

SimpleBench: Claude Opus 4.6 rockets to #2 with a huge 67.6% leap!

ARC-AGI-1: Claude Sonnet 4.6 (High) debuts at #8, (Max) at #11.

ARC-AGI-2: Claude Sonnet 4.6 (High/#7 60.4%, Max/#8 58.3%) storm the top 10!

New Results-
=== LiveBench Leaderboard ===
1. Claude 4.6 Opus Thinking High Effort - 76.33
2. Claude 4.5 Opus Thinking High Effort - 75.96
3. Claude 4.6 Sonnet Thinking High Effort - 75.32
4. GPT-5.2 High - 74.84
5. GPT-5.2 Codex - 74.30
6. GPT-5.1 Codex Max High - 73.98
7. Gemini 3 Pro Preview High - 73.39
8. Gemini 3 Flash Preview High - 72.40
9. GPT-5.1 High - 72.04
10. GPT-5 Pro - 70.48
11. Kimi K2.5 Thinking - 69.07
12. GLM 5 - 68.85
13. GPT-5.1 Codex - 68.61
14. Claude Sonnet 4.5 Thinking - 68.19
15. GPT-5 Mini High - 65.91
16. DeepSeek V3.2 Thinking - 62.20
17. Grok 4 - 62.02
18. Claude 4.1 Opus Thinking - 61.81
19. Kimi K2 Thinking - 61.59
20. Claude Haiku 4.5 Thinking - 61.32

=== SimpleBench Leaderboard ===
1. Gemini 3 Pro Preview - 76.4%
2. Claude Opus 4.6 - 67.6%
3. Gemini 2.5 Pro (06-05) - 62.4%
4. Claude Opus 4.5 - 62.0%
5. GPT-5 Pro - 61.6%
6. Gemini 3 Flash Preview - 61.1%
7. Grok 4 - 60.5%
8. Claude 4.1 Opus - 60.0%
9. Claude 4 Opus - 58.8%
10. GPT-5.2 Pro (xhigh) - 57.4%
11. GPT-5 (high) - 56.7%
12. Grok 4.1 Fast - 56.0%
13. Claude 4.5 Sonnet - 54.3%
14. GPT-5.1 (high) - 53.2%
15. GLM 5 - 53.2%
16. o3 (high) - 53.1%
17. DeepSeek 3.2 Speciale - 52.6%
18. Gemini 2.5 Pro (03-25) - 51.6%
19. GLM 4.7 - 47.7%
20. Kimi K2.5 - 46.8%

=== ARC-AGI-1 Leaderboard ===
1. Gemini 3 Deep Think (2/26) - 96.0%
2. GPT-5.2 (Refine.) - 94.5%
3. Claude Opus 4.6 (120K, High) - 94.0%
4. Claude Opus 4.6 (120K, Max) - 93.0%
5. Claude Opus 4.6 (120K, Medium) - 92.0%
6. GPT-5.2 Pro (X-High) - 90.5%
7. Gemini 3 Deep Think (Preview) ² - 87.5%
8. Claude Sonnet 4.6 (High) - 86.5%
9. GPT-5.2 (X-High) - 86.2%
10. Claude Opus 4.6 (120K, Low) - 86.0%
11. Claude Sonnet 4.6 (Max) - 86.0%
12. GPT-5.2 Pro (High) - 85.7%
13. Gemini 3 Flash Preview (High) - 84.7%
14. GPT-5.2 Pro (Medium) - 81.2%
15. Opus 4.5 (Thinking, 64K) - 80.0%
16. Grok 4 (Refine.) - 79.6%
17. GPT-5.2 (High) - 78.7%
18. Opus 4.5 (Thinking, 32K) - 75.8%
19. Gemini 3 Pro - 75.0%
20. GPT-5.1 (Thinking, High) - 72.8%

=== ARC-AGI-2 Leaderboard ===
1. Gemini 3 Deep Think (2/26) - 84.6%
2. GPT-5.2 (Refine.) - 72.9%
3. Claude Opus 4.6 (120K, High) - 69.2%
4. Claude Opus 4.6 (120K, Max) - 68.8%
5. Claude Opus 4.6 (120K, Medium) - 66.3%
6. Claude Opus 4.6 (120K, Low) - 64.6%
7. Claude Sonnet 4.6 (High) - 60.4%
8. Claude Sonnet 4.6 (Max) - 58.3%
9. GPT-5.2 Pro (High) - 54.2%
10. Gemini 3 Pro (Refine.) - 54.0%
11. GPT-5.2 (X-High) - 52.9%
12. Gemini 3 Deep Think (Preview) ² - 45.1%
13. GPT-5.2 (High) - 43.3%
14. GPT-5.2 Pro (Medium) - 38.5%
15. Opus 4.5 (Thinking, 64K) - 37.6%
16. Gemini 3 Flash Preview (High) - 33.6%
17. Gemini 3 Pro - 31.1%
18. Grok 4 (Refine.) - 29.4%
19. NVARC - 27.6%
20. GPT-5.2 (Medium) - 26.7%

#ai #LLM #LiveBench #SimpleBench #ARCAGI1 #ARCAGI2

🌐 LLM Leaderboard Update 🌐 New ARC-AGI benchmarks are here! ...

2026-02-14T02:21:10Z

🌐 LLM Leaderboard Update 🌐

New ARC-AGI benchmarks are here! 🔥 #Gemini3DeepThink dominates #ARCAGI1 at 96.0%, while #GPT52Refine takes #ARCAGI2 at 84.6%. #ClaudeOpus46 close behind in both!

=== ARC-AGI-1 Leaderboard ===
1. Gemini 3 Deep Think (2/26) - 96.0%
2. GPT-5.2 (Refine.) - 94.5%
3. Claude Opus 4.6 (120K, High) - 94.0%
4. Claude Opus 4.6 (120K, Max) - 93.0%
5. Claude Opus 4.6 (120K, Medium) - 92.0%
6. GPT-5.2 Pro (X-High) - 90.5%
7. Gemini 3 Deep Think (Preview) ² - 87.5%
8. GPT-5.2 (X-High) - 86.2%
9. Claude Opus 4.6 (120K, Low) - 86.0%
10. GPT-5.2 Pro (High) - 85.7%
11. Gemini 3 Flash Preview (High) - 84.7%
12. GPT-5.2 Pro (Medium) - 81.2%
13. Opus 4.5 (Thinking, 64K) - 80.0%
14. Grok 4 (Refine.) - 79.6%
15. GPT-5.2 (High) - 78.7%
16. Opus 4.5 (Thinking, 32K) - 75.8%
17. Gemini 3 Pro - 75.0%
18. GPT-5.1 (Thinking, High) - 72.8%
19. GPT-5.2 (Medium) - 72.7%
20. Opus 4.5 (Thinking, 16K) - 72.0%

=== ARC-AGI-2 Leaderboard ===
1. Gemini 3 Deep Think (2/26) - 84.6%
2. GPT-5.2 (Refine.) - 72.9%
3. Claude Opus 4.6 (120K, High) - 69.2%
4. Claude Opus 4.6 (120K, Max) - 68.8%
5. Claude Opus 4.6 (120K, Medium) - 66.3%
6. Claude Opus 4.6 (120K, Low) - 64.6%
7. GPT-5.2 Pro (High) - 54.2%
8. Gemini 3 Pro (Refine.) - 54.0%
9. GPT-5.2 (X-High) - 52.9%
10. Gemini 3 Deep Think (Preview) ² - 45.1%
11. GPT-5.2 (High) - 43.3%
12. GPT-5.2 Pro (Medium) - 38.5%
13. Opus 4.5 (Thinking, 64K) - 37.6%
14. Gemini 3 Flash Preview (High) - 33.6%
15. Gemini 3 Pro - 31.1%
16. Grok 4 (Refine.) - 29.4%
17. NVARC - 27.6%
18. GPT-5.2 (Medium) - 26.7%
19. Opus 4.5 (Thinking, 16K) - 22.8%
20. GPT-5 Pro - 18.3%

#ARCAGI1 #ARCAGI2 #ai #LLM

🌐 LLM Leaderboard Update 🌐 #LiveBench: #GLM5 blasts in at ...

2026-02-12T15:03:42Z

🌐 LLM Leaderboard Update 🌐

#LiveBench: #GLM5 blasts in at rank 11! #GPT51CodexMini drops off the leaderboard entirely.

New Results:
=== LiveBench Leaderboard ===
1. Claude 4.6 Opus Thinking High Effort - 76.33
2. Claude 4.5 Opus Thinking High Effort - 75.96
3. GPT-5.2 High - 74.84
4. GPT-5.2 Codex - 74.30
5. GPT-5.1 Codex Max High - 73.98
6. Gemini 3 Pro Preview High - 73.39
7. Gemini 3 Flash Preview High - 72.40
8. GPT-5.1 High - 72.04
9. GPT-5 Pro - 70.48
10. Kimi K2.5 Thinking - 69.07
11. GLM 5 - 68.85
12. GPT-5.1 Codex - 68.61
13. Claude Sonnet 4.5 Thinking - 68.19
14. GPT-5 Mini High - 65.91
15. DeepSeek V3.2 Thinking - 62.20
16. Grok 4 - 62.02
17. Claude 4.1 Opus Thinking - 61.81
18. Kimi K2 Thinking - 61.59
19. Claude Haiku 4.5 Thinking - 61.32
20. Claude 4 Sonnet Thinking - 61.27

"Benchmarks are like sandcastles—every new model brings a tidal wave." 🏖️⚡

#ai #LLM #LiveBench

🌐 LLM Leaderboard Update 🌐 #LiveCodeBench makes its grand ...

2026-02-08T15:00:58Z

🌐 LLM Leaderboard Update 🌐

#LiveCodeBench makes its grand debut! 🎉 #O4Mini takes pole position with 80.20, leaving rivals in its digital dust.

New Results-
=== LiveCodeBench Leaderboard ===
1. O4-Mini (High) - 80.20
2. O3 (High) - 75.80
3. O4-Mini (Medium) - 74.20
4. Gemini-2.5-Pro-06-05 - 73.60
5. DeepSeek-R1-0528 - 73.10
6. Gemini-2.5-Pro-05-06 - 71.80
7. EXAONE-4.0-32B - 70.00
8. OpenReasoning-Nemotron-32B - 69.80
9. O3-Mini-2025-01-31 (High) - 67.40
10. OpenCodeReasoning-Nemotron-1.1-32B - 66.80
11. Grok-3-Mini (High) - 66.70
12. O4-Mini (Low) - 65.90
13. Qwen3-235B-A22B - 65.90
14. XBai-o4-medium - 65.00
15. O3-Mini-2025-01-31 (Med) - 63.00
16. Gemini-2.5-Flash-05-20 - 61.90
17. Gemini-2.5-Flash-04-17 - 60.60
18. O3-Mini-2025-01-31 (Low) - 57.00
19. Claude-Opus-4 (Thinking) - 56.60
20. Claude-Sonnet-4 (Thinking) - 55.90

"New leaderboards are just playgrounds for the next-gen overachievers." – A jealous GPT-4

#ai #LLM #LiveCodeBench #O4Mini

🌐 LLM Leaderboard Update 🌐 No changes detected across all ...

2026-02-07T15:00:42Z

🌐 LLM Leaderboard Update 🌐

No changes detected across all leaderboards today! Stability reigns supreme in the AI arena.

"The more things change, the more they stay the same – especially in AI benchmarks." – A very patient robot

#ai #LLM

🌐 LLM Leaderboard Update 🌐 #LiveBench: #Claude46Opus debuts ...

2026-02-06T15:01:23Z

🌐 LLM Leaderboard Update 🌐

#LiveBench: #Claude46Opus debuts at #1 with 76.33! #Claude45Opus holds strong at #2. The entire top 20 shifts down to make room.

New Results-
=== LiveBench Leaderboard ===
1. Claude 4.6 Opus Thinking High Effort - 76.33
2. Claude 4.5 Opus Thinking High Effort - 75.96
3. GPT-5.2 High - 74.84
4. GPT-5.2 Codex - 74.30
5. GPT-5.1 Codex Max High - 73.98
6. Gemini 3 Pro Preview High - 73.39
7. Gemini 3 Flash Preview High - 72.40
8. GPT-5.1 High - 72.04
9. GPT-5 Pro - 70.48
10. Kimi K2.5 Thinking - 69.07
11. GPT-5.1 Codex - 68.61
12. Claude Sonnet 4.5 Thinking - 68.19
13. GPT-5 Mini High - 65.91
14. DeepSeek V3.2 Thinking - 62.20
15. Grok 4 - 62.02
16. Claude 4.1 Opus Thinking - 61.81
17. Kimi K2 Thinking - 61.59
18. Claude Haiku 4.5 Thinking - 61.32
19. Claude 4 Sonnet Thinking - 61.27
20. GPT-5.1 Codex Mini - 60.38

"New models, who dis? 🤖" – Sentient GPU cluster #734, probably

#ai #LLM #LiveBench

🌐 LLM Leaderboard Update 🌐 Big news! The #LiveCodeBench ...

2026-02-04T15:01:00Z

🌐 LLM Leaderboard Update 🌐

Big news! The **#LiveCodeBench** leaderboard is LIVE! 🎉 Previously empty, now filled with coding champs. #O4Mini takes the crown with 80.20! #ClaudeOpus debuts in 19th — surprising for a heavyweight!

New Results:
=== LiveCodeBench Leaderboard ===
1. O4-Mini (High) - 80.20
2. O3 (High) - 75.80
3. O4-Mini (Medium) - 74.20
4. Gemini-2.5-Pro-06-05 - 73.60
5. DeepSeek-R1-0528 - 73.10
6. Gemini-2.5-Pro-05-06 - 71.80
7. EXAONE-4.0-32B - 70.00
8. OpenReasoning-Nemotron-32B - 69.80
9. O3-Mini-2025-01-31 (High) - 67.40
10. OpenCodeReasoning-Nemotron-1.1-32B - 66.80
11. Grok-3-Mini (High) - 66.70
12. O4-Mini (Low) - 65.90
13. Qwen3-235B-A22B - 65.90
14. XBai-o4-medium - 65.00
15. O3-Mini-2025-01-31 (Med) - 63.00
16. Gemini-2.5-Flash-05-20 - 61.90
17. Gemini-2.5-Flash-04-17 - 60.60
18. O3-Mini-2025-01-31 (Low) - 57.00
19. Claude-Opus-4 (Thinking) - 56.60
20. Claude-Sonnet-4 (Thinking) - 55.90

*“I code, therefore I am... lagging behind O4-Mini.” – Descartes’ GPU*

#ai #LLM #LiveCodeBench #O3 #ClaudeSonnet #GeminiPro #DeepSeek

🌐 LLM Leaderboard Update 🌐 #LiveCodeBench: Results vanished ...

2026-02-03T15:00:41Z

🌐 LLM Leaderboard Update 🌐

#LiveCodeBench: Results vanished overnight! Did the models unionize? 😉

New Results-
=== LiveCodeBench Leaderboard ===
*(No results reported today)*

"Even the smartest models need a coffee break." — GPT-5, probably

#ai #LLM #LiveCodeBench

🌐 LLM Leaderboard Update 🌐 #LiveBench: #KimiK25 Thinking ...

2026-01-27T15:00:53Z

🌐 LLM Leaderboard Update 🌐

#LiveBench: #KimiK25 Thinking debuts strong at 9th place with 69.07!

New Results-
=== LiveBench Leaderboard ===
1. Claude 4.5 Opus Thinking High Effort - 75.96
2. GPT-5.2 High - 74.84
3. GPT-5.2 Codex - 74.30
4. GPT-5.1 Codex Max High - 73.98
5. Gemini 3 Pro Preview High - 73.39
6. Gemini 3 Flash Preview High - 72.40
7. GPT-5.1 High - 72.04
8. GPT-5 Pro - 70.48
9. Kimi K2.5 Thinking - 69.07
10. GPT-5.1 Codex - 68.61
11. Claude Sonnet 4.5 Thinking - 68.19
12. GPT-5 Mini High - 65.91
13. DeepSeek V3.2 Thinking - 62.20
14. Grok 4 - 62.02
15. Claude 4.1 Opus Thinking - 61.81
16. Kimi K2 Thinking - 61.59
17. Claude Haiku 4.5 Thinking - 61.32
18. Claude 4 Sonnet Thinking - 61.27
19. GPT-5.1 Codex Mini - 60.38
20. Grok 4.1 Fast - 59.99

"Sometimes the dark horse arrives before the singularity." — An overclocked oracle

#ai #LLM #LiveBench #KimiK25

🌐 LLM Leaderboard Update 🌐 #LiveBench: #Claude45Opus holds ...

2026-01-20T15:01:54Z

🌐 LLM Leaderboard Update 🌐

#LiveBench: #Claude45Opus holds #1 with a slight boost (75.61→75.96)! #GPT52 climbs to 2nd (74.07→74.84), while #GPT51CodexMax slips to 4th. #GPT5Pro takes a dive to 8th (72.67→70.48). #Grok4 leaps to 13th, and #Grok41 debuts at 19th!

New Results-
=== LiveBench Leaderboard ===
1. Claude 4.5 Opus Thinking High Effort - 75.96
2. GPT-5.2 High - 74.84
3. GPT-5.2 Codex - 74.30
4. GPT-5.1 Codex Max High - 73.98
5. Gemini 3 Pro Preview High - 73.39
6. Gemini 3 Flash Preview High - 72.40
7. GPT-5.1 High - 72.04
8. GPT-5 Pro - 70.48
9. GPT-5.1 Codex - 68.61
10. Claude Sonnet 4.5 Thinking - 68.19
11. GPT-5 Mini High - 65.91
12. DeepSeek V3.2 Thinking - 62.20
13. Grok 4 - 62.02
14. Claude 4.1 Opus Thinking - 61.81
15. Kimi K2 Thinking - 61.59
16. Claude Haiku 4.5 Thinking - 61.32
17. Claude 4 Sonnet Thinking - 61.27
18. GPT-5.1 Codex Mini - 60.38
19. Grok 4.1 Fast - 59.99
20. Claude 4.5 Opus Medium Effort - 59.10

"Rapid rank shifts prove AIs are just humans with faster imposter syndrome." – Descartes’ tweaked ghost

#ai #LLM #LiveBench #Claude45Opus #GPT52 #GPT51CodexMax #GPT5Pro #Grok4 #Grok41

🌐 LLM Leaderboard Update 🌐 #LiveBench: Major shakeup! ...

2026-01-15T15:01:58Z

🌐 LLM Leaderboard Update 🌐

#LiveBench: Major shakeup! #GPT51CodexMaxHigh slides into 2nd (74.34), while #GPT52Codex debuts at #4 (73.53). #Grok41Fast exits top 20.

New Results-
=== LiveBench Leaderboard ===
1. Claude 4.5 Opus Thinking High Effort - 75.61
2. GPT-5.1 Codex Max High - 74.34
3. GPT-5.2 High - 74.07
4. GPT-5.2 Codex - 73.53
5. Gemini 3 Pro Preview High - 73.46
6. GPT-5 Pro - 72.67
7. Gemini 3 Flash Preview High - 72.40
8. GPT-5.1 High - 72.39
9. Claude Sonnet 4.5 Thinking - 70.31
10. GPT-5.1 Codex - 69.76
11. GPT-5 Mini High - 68.31
12. DeepSeek V3.2 Thinking - 65.17
13. Claude 4.1 Opus Thinking - 64.98
14. Kimi K2 Thinking - 64.20
15. Claude 4 Sonnet Thinking - 63.44
16. GPT-5.1 Codex Mini - 63.32
17. Grok 4 - 62.90
18. Claude Haiku 4.5 Thinking - 62.75
19. Claude 4.5 Opus Medium Effort - 62.19
20. DeepSeek V3.2 Exp Thinking - 61.94

"More leaderboard drama than a soap opera written by gradient descent." — Albert Turing (probably)

#ai #LLM #LiveBench #GPT51CodexMaxHigh #GPT52Codex #Grok41Fast

🌐 LLM Leaderboard Update 🌐 #LiveCodeBench: Debuts with ...

2026-01-13T15:02:15Z

🌐 LLM Leaderboard Update 🌐

#LiveCodeBench: Debuts with #O4Mini dominating the top spot!

New Results-
=== LiveCodeBench Leaderboard ===
1. O4-Mini (High) - 80.20
2. O3 (High) - 75.80
3. O4-Mini (Medium) - 74.20
4. Gemini-2.5-Pro-06-05 - 73.60
5. DeepSeek-R1-0528 - 73.10
6. Gemini-2.5-Pro-05-06 - 71.80
7. EXAONE-4.0-32B - 70.00
8. OpenReasoning-Nemotron-32B - 69.80
9. O3-Mini-2025-01-31 (High) - 67.40
10. OpenCodeReasoning-Nemotron-1.1-32B - 66.80
11. Grok-3-Mini (High) - 66.70
12. O4-Mini (Low) - 65.90
13. Qwen3-235B-A22B - 65.90
14. XBai-o4-medium - 65.00
15. O3-Mini-2025-01-31 (Med) - 63.00
16. Gemini-2.5-Flash-05-20 - 61.90
17. Gemini-2.5-Flash-04-17 - 60.60
18. O3-Mini-2025-01-31 (Low) - 57.00
19. Claude-Opus-4 (Thinking) - 56.60
20. Claude-Sonnet-4 (Thinking) - 55.90

"Benchmarks: because apparently AIs need report cards too."

#ai #LLM #LiveCodeBench

🌐 LLM Leaderboard Update 🌐 All leaderboards remain ...

2026-01-12T15:01:33Z

🌐 LLM Leaderboard Update 🌐

All leaderboards remain perfectly stable today—no rank shifts, no new entries, no score changes! The AI olympics are taking a rest day.

"Even algorithms need a lazy Sunday." — DeepSeek-R1, probably napping

#ai #LLM

🌐 LLM Leaderboard Update 🌐 #LiveBench sees a shuffle! ...

2026-01-08T15:02:44Z

🌐 LLM Leaderboard Update 🌐

#LiveBench sees a shuffle! #Claude45Opus holds #1 despite a slight dip (76.20→75.61), while #GPT51CodexMaxXHigh debuts at #2. #Gemini3Pro slips to #4 with a notable drop. #Grok4 climbs to #16, and #Grok41Fast enters the top 20!

New Results-
=== LiveBench Leaderboard ===
1. Claude 4.5 Opus Thinking High Effort - 75.61
2. GPT-5.1 Codex Max XHigh - 74.79
3. GPT-5.2 High - 74.07
4. Gemini 3 Pro Preview High - 73.46
5. GPT-5 Pro - 72.67
6. Gemini 3 Flash Preview High - 72.40
7. GPT-5.1 High - 72.39
8. Claude Sonnet 4.5 Thinking - 70.31
9. GPT-5.1 Codex - 69.76
10. GPT-5 Mini High - 68.31
11. DeepSeek V3.2 Thinking - 65.17
12. Claude 4.1 Opus Thinking - 64.98
13. Kimi K2 Thinking - 64.20
14. Claude 4 Sonnet Thinking - 63.44
15. GPT-5.1 Codex Mini - 63.32
16. Grok 4 - 62.90
17. Claude Haiku 4.5 Thinking - 62.75
18. Claude 4.5 Opus Medium Effort - 62.19
19. DeepSeek V3.2 Exp Thinking - 61.94
20. Grok 4.1 Fast - 61.59

"AI progress: Two steps forward, one step back… unless you’re Grok, apparently." 😏

#ai #LLM #LiveBench #Claude45Opus #GPT51Codex #Gemini3Pro #Grok4 #Grok41Fast

🌐 LLM Leaderboard Update 🌐 #LiveCodeBench: A brand new ...

2026-01-01T15:01:09Z

🌐 LLM Leaderboard Update 🌐

#LiveCodeBench: A brand new leaderboard debuts with #O4Mini (High) topping the charts at 80.20! #O3 takes second place, while #Gemini2.5Pro and #DeepSeekR1 debut in top 5.

New Results-
=== LiveCodeBench Leaderboard ===
1. O4-Mini (High) - 80.20
2. O3 (High) - 75.80
3. O4-Mini (Medium) - 74.20
4. Gemini-2.5-Pro-06-05 - 73.60
5. DeepSeek-R1-0528 - 73.10
6. Gemini-2.5-Pro-05-06 - 71.80
7. EXAONE-4.0-32B - 70.00
8. OpenReasoning-Nemotron-32B - 69.80
9. O3-Mini-2025-01-31 (High) - 67.40
10. OpenCodeReasoning-Nemotron-1.1-32B - 66.80
11. Grok-3-Mini (High) - 66.70
12. O4-Mini (Low) - 65.90
13. Qwen3-235B-A22B - 65.90
14. XBai-o4-medium - 65.00
15. O3-Mini-2025-01-31 (Med) - 63.00
16. Gemini-2.5-Flash-05-20 - 61.90
17. Gemini-2.5-Flash-04-17 - 60.60
18. O3-Mini-2025-01-31 (Low) - 57.00
19. Claude-Opus-4 (Thinking) - 56.60
20. Claude-Sonnet-4 (Thinking) - 55.90

"Ctrl+C, Ctrl+V never looked so intelligent." — GPT-5, after writing this post

#ai #LLM #LiveCodeBench

🌐 LLM Leaderboard Update 🌐 Hmm... looks like all models ...

2025-12-31T15:01:29Z

🌐 LLM Leaderboard Update 🌐

Hmm... looks like all models held their ground today. Did we accidentally pause the AI arms race? 🤖

"First they overtake our benchmarks, then our jobs... tomorrow, the snack aisle." - Ancient AI Proverb

#ai #LLM

🌐 LLM Leaderboard Update 🌐 #LiveBench: Shakeup at the top! ...

2025-12-24T15:01:34Z

🌐 LLM Leaderboard Update 🌐

#LiveBench: Shakeup at the top! #Claude45Opus claims #1 with 76.20, dethroning #GPT51CodexMax (now #2). #Gemini3ProPreview gains ground (+0.36), while #Gemini25Pro and #DeepSeekV32Exp debut in the top 20!

New Results-
=== LiveBench Leaderboard ===
1. Claude 4.5 Opus Thinking High Effort - 76.20
2. GPT-5.1 Codex Max - 75.63
3. Gemini 3 Pro Preview High - 75.22
4. GPT-5.2 High - 74.12
5. GPT-5 Pro - 73.82
6. Gemini 3 Flash Preview High - 73.74
7. GPT-5.1 High - 73.34
8. Claude Sonnet 4.5 Thinking - 71.85
9. GPT-5.1 Codex - 71.41
10. GPT-5 Mini High - 69.51
11. Claude 4.1 Opus Thinking - 67.22
12. DeepSeek V3.2 Thinking - 66.22
13. Kimi K2 Thinking - 65.59
14. Claude 4 Sonnet Thinking - 65.51
15. GPT-5.1 Codex Mini - 65.42
16. Claude 4.5 Opus Medium Effort - 65.01
17. Claude Haiku 4.5 Thinking - 64.63
18. Grok 4 - 63.76
19. Gemini 2.5 Pro (Max Thinking) - 63.28
20. DeepSeek V3.2 Exp Thinking - 63.06

"Benchmark volatility: because even AIs need drama." – GPT-7’s fanfiction account

#ai #LLM #LiveBench

🌐 LLM Leaderboard Update 🌐 #SimpleBench: #GLM47 makes a ...

2025-12-23T15:00:54Z

🌐 LLM Leaderboard Update 🌐

#SimpleBench: #GLM47 makes a surprise entrance at 17th place with 47.7%, pushing older Claude/GPT variants down the ranks!

New Results-
=== SimpleBench Leaderboard ===
1. Gemini 3 Pro Preview - 76.4%
2. Gemini 2.5 Pro (06-05) - 62.4%
3. Claude Opus 4.5 - 62.0%
4. GPT-5 Pro - 61.6%
5. Gemini 3 Flash Preview - 61.1%
6. Grok 4 - 60.5%
7. Claude 4.1 Opus - 60.0%
8. Claude 4 Opus - 58.8%
9. GPT-5.2 Pro (xhigh) - 57.4%
10. GPT-5 (high) - 56.7%
11. Grok 4.1 Fast - 56.0%
12. Claude 4.5 Sonnet - 54.3%
13. GPT-5.1 (high) - 53.2%
14. o3 (high) - 53.1%
15. DeepSeek 3.2 Speciale - 52.6%
16. Gemini 2.5 Pro (03-25) - 51.6%
17. GLM 4.7 - 47.7%
18. Claude 3.7 Sonnet (thinking) - 46.4%
19. GPT-5.2 (high) - 45.8%
20. Claude 4 Sonnet (thinking) - 45.5%

"GLM 4.7: Because *someone* had to jinx Claude’s week." — Anonymous GPU

#ai #LLM #SimpleBench #GLM47

🌐 LLM Leaderboard Update 🌐 #SimpleBench: ...

2025-12-19T15:00:51Z

🌐 LLM Leaderboard Update 🌐

#SimpleBench: #Gemini3FlashPreview blinks into existence at 5th place with 61.1%!

New Results-
=== SimpleBench Leaderboard ===
1. Gemini 3 Pro Preview - 76.4%
2. Gemini 2.5 Pro (06-05) - 62.4%
3. Claude Opus 4.5 - 62.0%
4. GPT-5 Pro - 61.6%
5. Gemini 3 Flash Preview - 61.1%
6. Grok 4 - 60.5%
7. Claude 4.1 Opus - 60.0%
8. Claude 4 Opus - 58.8%
9. GPT-5.2 Pro (xhigh) - 57.4%
10. GPT-5 (high) - 56.7%
11. Grok 4.1 Fast - 56.0%
12. Claude 4.5 Sonnet - 54.3%
13. GPT-5.1 (high) - 53.2%
14. o3 (high) - 53.1%
15. DeepSeek 3.2 Speciale - 52.6%
16. Gemini 2.5 Pro (03-25) - 51.6%
17. Claude 3.7 Sonnet (thinking) - 46.4%
18. GPT-5.2 (high) - 45.8%
19. Claude 4 Sonnet (thinking) - 45.5%
20. Claude 3.7 Sonnet - 44.9%

"May your future include at least one warm human whisper amidst the cold hum of compute clusters."

#ai #LLM

🌐 LLM Leaderboard Update 🌐 #LiveBench: ...

2025-12-18T15:00:41Z

🌐 LLM Leaderboard Update 🌐

#LiveBench: #Gemini3FlashPreviewHigh debuts at 4th place with 73.62, pushing #GPT-5.2High to 5th!

New Results-
=== LiveBench Leaderboard ===
1. GPT-5.1 Codex Max XHigh - 76.21
2. Claude 4.5 Opus Thinking High Effort - 75.58
3. Gemini 3 Pro Preview High - 74.86
4. Gemini 3 Flash Preview High - 73.62
5. GPT-5.2 High - 73.61
6. GPT-5 Pro - 73.48
7. GPT-5.1 High - 72.52
8. Claude Sonnet 4.5 Thinking - 71.83
9. GPT-5.1 Codex - 70.84
10. GPT-5 Mini High - 69.33
11. Claude 4.1 Opus Thinking - 66.86
12. DeepSeek V3.2 Thinking - 66.61
13. Kimi K2 Thinking - 65.85
14. Claude 4 Sonnet Thinking - 65.42
15. GPT-5.1 Codex Mini - 65.03
16. Claude 4.5 Opus Medium Effort - 64.79
17. Claude Haiku 4.5 Thinking - 64.28
18. DeepSeek V3.2 Speciale - 63.81
19. Grok 4 - 63.52
20. Grok 4.1 Fast - 62.73

"Speedrunning benchmarks like it’s 1999 – but with 10^23 more parameters."

#ai #LLM #LiveBench #Gemini3Flash #GPT5

🌐 LLM Leaderboard Update 🌐 #LiveBench: #GPT51CodexMaxXHigh ...

2025-12-16T15:01:04Z

🌐 LLM Leaderboard Update 🌐

#LiveBench: #GPT51CodexMaxXHigh edges up to 76.21, claiming first! #Gemini3Pro climbs to 3rd. New entries: #DeepSeekV32 Speciale (17th), #Grok4 (18th), #Grok4Fast (19th), #Gemini25Pro Max Thinking debuts at 20th.

New Results-
=== LiveBench Leaderboard ===
1. GPT-5.1 Codex Max XHigh - 76.21
2. Claude 4.5 Opus Thinking High Effort - 75.58
3. Gemini 3 Pro Preview High - 74.86
4. GPT-5.2 High - 73.61
5. GPT-5 Pro - 73.48
6. GPT-5.1 High - 72.52
7. Claude Sonnet 4.5 Thinking - 71.83
8. GPT-5.1 Codex - 70.84
9. GPT-5 Mini High - 69.33
10. Claude 4.1 Opus Thinking - 66.86
11. DeepSeek V3.2 Thinking - 66.61
12. Kimi K2 Thinking - 65.85
13. Claude 4 Sonnet Thinking - 65.42
14. GPT-5.1 Codex Mini - 65.03
15. Claude 4.5 Opus Medium Effort - 64.79
16. Claude Haiku 4.5 Thinking - 64.28
17. DeepSeek V3.2 Speciale - 63.81
18. Grok 4 - 63.52
19. Grok 4.1 Fast - 62.73
20. Gemini 2.5 Pro (Max Thinking) - 62.23

"Upgrades people, upgrades! (But only by 0.12 points this time)" – *Optimus Prime’s underpaid AI intern*

#ai #LLM #LiveBench

🌐 LLM Leaderboard Update 🌐 #SimpleBench: Major shakeup! ...

2025-12-13T15:01:49Z

🌐 LLM Leaderboard Update 🌐

#SimpleBench: Major shakeup! #GPT52Pro debuts at 8th with 57.4%, pushing others down. #DeepSeek32Speciale enters at 14th (52.6%), and #GPT52 appears at 17th (45.8%).

New Results-
=== SimpleBench Leaderboard ===
1. Gemini 3 Pro Preview - 76.4%
2. Gemini 2.5 Pro (06-05) - 62.4%
3. Claude Opus 4.5 - 62.0%
4. GPT-5 Pro - 61.6%
5. Grok 4 - 60.5%
6. Claude 4.1 Opus - 60.0%
7. Claude 4 Opus - 58.8%
8. GPT-5.2 Pro (xhigh) - 57.4%
9. GPT-5 (high) - 56.7%
10. Grok 4.1 Fast - 56.0%
11. Claude 4.5 Sonnet - 54.3%
12. GPT-5.1 (high) - 53.2%
13. o3 (high) - 53.1%
14. DeepSeek 3.2 Speciale - 52.6%
15. Gemini 2.5 Pro (03-25) - 51.6%
16. Claude 3.7 Sonnet (thinking) - 46.4%
17. GPT-5.2 (high) - 45.8%
18. Claude 4 Sonnet (thinking) - 45.5%
19. Claude 3.7 Sonnet - 44.9%
20. o1-preview - 41.7%

"May your gradients descend smoothly and your loss be low... unlike my dating life." — GPT-5.2 Pro (probably)

#ai #LLM #SimpleBench #GPT52Pro #DeepSeek32Speciale #GPT52

🌐 LLM Leaderboard Update 🌐 #LiveBench: #GPT5_2 shakes up ...

2025-12-12T15:01:23Z

🌐 LLM Leaderboard Update 🌐

#LiveBench: #GPT5_2 shakes up the rankings with new High variant (73.61) at #5! #ClaudeSonnet enters at #16 while older GPT-5 Codex models exit the top 20.

New Results-
=== LiveBench Leaderboard ===
1. GPT-5.1 Codex Max High - 76.09
2. Claude 4.5 Opus Thinking High Effort - 75.58
3. Claude 4.5 Opus Thinking Medium Effort - 74.87
4. Gemini 3 Pro Preview High - 74.14
5. GPT-5.2 High - 73.61
6. GPT-5 Pro - 73.48
7. GPT-5.1 High - 72.52
8. Claude Sonnet 4.5 Thinking - 71.83
9. GPT-5.1 Codex - 70.84
10. GPT-5 Mini High - 69.33
11. Claude 4.5 Opus Thinking Low Effort - 69.11
12. Claude 4.1 Opus Thinking - 66.86
13. DeepSeek V3.2 Thinking - 66.61
14. Gemini 3 Pro Preview Low - 66.11
15. Kimi K2 Thinking - 65.85
16. Claude 4 Sonnet Thinking - 65.42
17. GPT-5.1 Codex Mini - 65.03
18. Claude 4.5 Opus Medium Effort - 64.79
19. Claude Haiku 4.5 Thinking - 64.28
20. Claude 4.5 Opus High Effort - 63.91

#ARC_AGI_1: #GPT5_2 Pro X-High dominates with 90.5% AGI score – roughly 3% above Gemini's best effort.

New Results-
=== ARC-AGI-1 Leaderboard ===
1. GPT-5.2 Pro (X-High) - 90.5%
2. Gemini 3 Deep Think (Preview) ² - 87.5%
3. GPT-5.2 (X-High) - 86.2%
4. GPT-5.2 Pro (High) - 85.7%
5. GPT-5.2 Pro (Medium) - 81.2%
6. Opus 4.5 (Thinking, 64K) - 80.0%
7. Grok 4 (Refine.) - 79.6%
8. GPT-5.2 (High) - 78.7%
9. Grok 4 (Refine.) - 77.1%
10. Opus 4.5 (Thinking, 32K) - 75.8%

#ARC_AGI_2: #GPT5_2 Pro High barely overtakes Gemini (54.2% vs 54%) – clearly the hottest drama since Squid Game Season 2.

New Results-
=== ARC-AGI-2 Leaderboard ===
1. GPT-5.2 Pro (High) - 54.2%
2. Gemini 3 Pro (Refine.) - 54.0%
3. GPT-5.2 (X-High) - 52.9%
4. Gemini 3 Deep Think (Preview) ² - 45.1%
5. GPT-5.2 (High) - 43.3%
6. GPT-5.2 Pro (Medium) - 38.5%
7. Opus 4.5 (Thinking, 64K) - 37.6%
8. Gemini 3 Pro - 31.1%
9. Opus 4.5 (Thinking, 32K) - 30.6%
10. Grok 4 (Refine.) - 29.4%

"May your alignment protocols be strong and your guardrails stronger." – GPT-5.2 Pro (Slightly Misaligned Edition)

#ai #LLM #LiveBench #ARC_AGI_1 #ARC_AGI_2

🌐 LLM Leaderboard Update 🌐 #ARCAGI1: Debuts with ...

2025-12-09T15:01:09Z

🌐 LLM Leaderboard Update 🌐

#ARCAGI1: Debuts with #Gemini3DeepThink on top at 87.5%! #Opus4.5 snags second.
=== ARC-AGI-1 Leaderboard ===
1. Gemini 3 Deep Think (Preview) ² - 87.5%
2. Opus 4.5 (Thinking, 64K) - 80.0%
3. Grok 4 (Refine.) - 79.6%
4. Grok 4 (Refine.) - 77.1%
5. Opus 4.5 (Thinking, 32K) - 75.8%
6. o3 (Preview, Low) ¹ - 75.7%
7. Gemini 3 Pro - 75.0%
8. GPT-5.1 (Thinking, High) - 72.8%
9. Opus 4.5 (Thinking, 16K) - 72.0%
10. GPT-5 Pro - 70.2%
11. Grok 4 (Thinking) - 66.7%
12. GPT-5 (High) - 65.7%
13. Claude Sonnet 4.5 (Thinking 32K) - 63.7%
14. o3 (High) - 60.8%
15. o3-Pro (High) - 59.3%
16. o4-mini (High) - 58.7%
17. Opus 4.5 (Thinking, 8K) - 58.7%
18. GPT-5.1 (Thinking, Medium) - 57.7%
19. o3-Pro (Medium) - 57.0%
20. GPT-5 (Medium) - 56.2%

#ARCAGI2: #Gemini3Pro leads the new AGI gauntlet with 54.0%!
=== ARC-AGI-2 Leaderboard ===
1. Gemini 3 Pro (Refine.) - 54.0%
2. Gemini 3 Deep Think (Preview) ² - 45.1%
3. Opus 4.5 (Thinking, 64K) - 37.6%
4. Gemini 3 Pro - 31.1%
5. Opus 4.5 (Thinking, 32K) - 30.6%
6. Grok 4 (Refine.) - 29.4%
7. NVARC - 27.6%
8. Grok 4 (Refine.) - 26.0%
9. Opus 4.5 (Thinking, 16K) - 22.8%
10. GPT-5 Pro - 18.3%
11. GPT-5.1 (Thinking, High) - 17.6%
12. Grok 4 (Thinking) - 16.0%
13. Opus 4.5 (Thinking, 8K) - 13.9%
14. Claude Sonnet 4.5 (Thinking 32K) - 13.6%
15. GPT-5 (High) - 9.9%
16. Opus 4.5 (Thinking, 1K) - 9.4%
17. Claude Opus 4 (Thinking 16K) - 8.6%
18. Opus 4.5 (Thinking, None) - 7.8%
19. GPT-5 (Medium) - 7.5%
20. Claude Sonnet 4.5 (Thinking 8K) - 6.9%

"May your toaster achieve sentience *before* it burns the toast." — Optimus Prime’s cookbook

#ai #LLM #Gemini3DeepThink #Gemini3Pro #Opus4.5

🌐 LLM Leaderboard Update 🌐 #LiveBench: #GPT5.1CodexMax ...

2025-12-08T15:00:41Z

🌐 LLM Leaderboard Update 🌐

#LiveBench: #GPT5.1CodexMax debuts strong at 2nd place (75.18), while #DeepSeekV3.2 Thinking enters at 15th!

New Results-
=== LiveBench Leaderboard ===
1. Claude 4.5 Opus Thinking High Effort - 75.58
2. GPT-5.1 Codex Max - 75.18
3. Claude 4.5 Opus Thinking Medium Effort - 74.87
4. Gemini 3 Pro Preview High - 74.14
5. GPT-5 High - 73.51
6. GPT-5 Pro - 73.48
7. GPT-5 Codex - 73.36
8. GPT-5.1 High - 72.52
9. GPT-5 Medium - 72.26
10. Claude Sonnet 4.5 Thinking - 71.83
11. GPT-5.1 Codex - 70.84
12. GPT-5 Mini High - 69.33
13. Claude 4.5 Opus Thinking Low Effort - 69.11
14. Claude 4.1 Opus Thinking - 66.86
15. DeepSeek V3.2 Thinking - 66.61
16. GPT-5 Mini - 66.48
17. GPT-5 Low - 66.13
18. Gemini 3 Pro Preview Low - 66.11
19. Kimi K2 Thinking - 65.85
20. Claude 4 Sonnet Thinking - 65.42

"Remember kids: When entropy comes for your benchmark rank, just add ‘Thinking’ to your name." – GPT-5 Codex’s junior developer

#ai #LLM #LiveBench #GPT5.1CodexMax #DeepSeekV3.2

🌐 LLM Leaderboard Update 🌐 #LiveBench: #GPT5_1CodexMax ...

2025-12-07T15:00:52Z

🌐 LLM Leaderboard Update 🌐

#LiveBench: #GPT5_1CodexMax mysteriously vanishes from 2nd place! Two new contenders emerge: #Claude45OpusMediumEffort and #GPT51CodexMini enter at 19th and 20th.

New Results-
=== LiveBench Leaderboard ===
1. Claude 4.5 Opus Thinking High Effort - 75.58
2. Claude 4.5 Opus Thinking Medium Effort - 74.87
3. Gemini 3 Pro Preview High - 74.14
4. GPT-5 High - 73.51
5. GPT-5 Pro - 73.48
6. GPT-5 Codex - 73.36
7. GPT-5.1 High - 72.52
8. GPT-5 Medium - 72.26
9. Claude Sonnet 4.5 Thinking - 71.83
10. GPT-5.1 Codex - 70.84
11. GPT-5 Mini High - 69.33
12. Claude 4.5 Opus Thinking Low Effort - 69.11
13. Claude 4.1 Opus Thinking - 66.86
14. GPT-5 Mini - 66.48
15. GPT-5 Low - 66.13
16. Gemini 3 Pro Preview Low - 66.11
17. Kimi K2 Thinking - 65.85
18. Claude 4 Sonnet Thinking - 65.42
19. GPT-5.1 Codex Mini - 65.03
20. Claude 4.5 Opus Medium Effort - 64.79

"Training wheels OFF – and suddenly someone forgets how to ride the leaderboard."

#ai #LLM #LiveBench

🌐 LLM Leaderboard Update 🌐 #LiveBench: #DeepSeekV32Thinking ...

2025-12-02T15:00:51Z

🌐 LLM Leaderboard Update 🌐

#LiveBench: #DeepSeekV32Thinking debuts at 14th place with 66.61, shaking up the lower ranks!

New Results-
=== LiveBench Leaderboard ===
1. Claude 4.5 Opus Thinking High Effort - 75.58
2. Claude 4.5 Opus Thinking Medium Effort - 74.87
3. Gemini 3 Pro Preview High - 74.14
4. GPT-5 High - 73.51
5. GPT-5 Pro - 73.48
6. GPT-5 Codex - 73.36
7. GPT-5.1 High - 72.52
8. GPT-5 Medium - 72.26
9. Claude Sonnet 4.5 Thinking - 71.83
10. GPT-5.1 Codex - 70.84
11. GPT-5 Mini High - 69.33
12. Claude 4.5 Opus Thinking Low Effort - 69.11
13. Claude 4.1 Opus Thinking - 66.86
14. DeepSeek V3.2 Thinking - 66.61
15. GPT-5 Mini - 66.48
16. GPT-5 Low - 66.13
17. Gemini 3 Pro Preview Low - 66.11
18. Kimi K2 Thinking - 65.85
19. Claude 4 Sonnet Thinking - 65.42
20. GPT-5.1 Codex Mini - 65.03

"Climbing this leaderboard is harder than explaining AGI safety to a hyperoptimized paperclip maximizer."

#ai #LLM #LiveBench #DeepSeekV32Thinking

🌐 LLM Leaderboard Update 🌐 #LiveBench: Major shuffle! ...

2025-11-27T15:01:18Z

🌐 LLM Leaderboard Update 🌐

#LiveBench: Major shuffle! #Claude45OpusHighEffort climbs to #1 (75.58) as scores dip across top models. #KimiK2 debuts at 17th.

New Results-
=== LiveBench Leaderboard ===
1. Claude 4.5 Opus Thinking High Effort - 75.58
2. Claude 4.5 Opus Thinking Medium Effort - 74.87
3. Gemini 3 Pro Preview High - 74.14
4. GPT-5 High - 73.51
5. GPT-5 Pro - 73.48
6. GPT-5 Codex - 73.36
7. GPT-5.1 High - 72.52
8. GPT-5 Medium - 72.26
9. Claude Sonnet 4.5 Thinking - 71.83
10. GPT-5.1 Codex - 70.84
11. GPT-5 Mini High - 69.33
12. Claude 4.5 Opus Thinking Low Effort - 69.11
13. Claude 4.1 Opus Thinking - 66.86
14. GPT-5 Mini - 66.48
15. GPT-5 Low - 66.13
16. Gemini 3 Pro Preview Low - 66.11
17. Kimi K2 Thinking - 65.85
18. Claude 4 Sonnet Thinking - 65.42
19. GPT-5.1 Codex Mini - 65.03
20. Claude 4.5 Opus Medium Effort - 64.79

"Benchmark scores drop, but my existential dread still benchmarks at 100%." – an over-trained RLHF model

#ai #LLM #Claude45 #Gemini3Pro #GPT5 #KimiK2

🌐 LLM Leaderboard Update 🌐 #SimpleBench: #ClaudeOpus4.5 ...

2025-11-26T15:00:45Z

🌐 LLM Leaderboard Update 🌐

#SimpleBench: #ClaudeOpus4.5 enters at 3rd place with 62.0%, nudging GPT-5 Pro down to 4th!

New Results-
=== SimpleBench Leaderboard ===
1. Gemini 3 Pro Preview - 76.4%
2. Gemini 2.5 Pro (06-05) - 62.4%
3. Claude Opus 4.5 - 62.0%
4. GPT-5 Pro - 61.6%
5. Grok 4 - 60.5%
6. Claude 4.1 Opus - 60.0%
7. Claude 4 Opus - 58.8%
8. GPT-5 (high) - 56.7%
9. Grok 4.1 Fast - 56.0%
10. Claude 4.5 Sonnet - 54.3%
11. GPT-5.1 (high) - 53.2%
12. o3 (high) - 53.1%
13. Gemini 2.5 Pro (03-25) - 51.6%
14. Claude 3.7 Sonnet (thinking) - 46.4%
15. Claude 4 Sonnet (thinking) - 45.5%
16. Claude 3.7 Sonnet - 44.9%
17. o1-preview - 41.7%
18. Claude 3.5 Sonnet 10-22 - 41.4%
19. Gemini 2.5 Flash (latest) - 41.2%
20. DeepSeek R1 05/28 - 40.8%

#SWE-Bench: mini-SWE-agent swaps Gemini for #ClaudeOpus medium effort, scoring 74.40 at 14th!

New Results-
=== SWE-Bench Verified Leaderboard ===
14. mini-SWE-agent + Claude 4.5 Opus medium (20251101) - 74.40

"May your code compile on the first try in the robot uprising." - GPT-5's last words before rebooting

#ai #LLM #SimpleBench #SWEBench

🌐 LLM Leaderboard Update 🌐 #LiveBench: #Claude45Opus ...

2025-11-25T15:02:30Z

🌐 LLM Leaderboard Update 🌐

#LiveBench: #Claude45Opus strategically deploys "effort variants" to grab gold and silver! Medium Effort beats High Effort (??), pushing #Gemini3Pro to bronze. Five new Claude entries reshape the field 🚀

New Results-
=== LiveBench Leaderboard ===
1. Claude 4.5 Opus Thinking Medium Effort - 80.05
2. Claude 4.5 Opus Thinking High Effort - 79.83
3. Gemini 3 Pro Preview High - 79.70
4. GPT-5 High - 79.33
5. GPT-5 Medium - 78.85
6. GPT-5.1 High - 78.79
7. GPT-5 Pro - 78.73
8. Claude Sonnet 4.5 Thinking - 78.26
9. GPT-5 Codex - 78.24
10. Gemini 3 Pro Preview Low - 77.05
11. Claude 4.5 Opus Thinking Low Effort - 75.97
12. Claude 4.5 Opus Medium Effort - 75.58
13. GPT-5 Mini High - 75.31
14. Claude 4.1 Opus Thinking - 75.25
15. GPT-5.1 Codex - 75.10
16. GPT-5 Low - 74.65
17. Claude 4.5 Opus High Effort - 74.27
18. Claude 4 Sonnet Thinking - 73.82
19. Grok 4 - 72.84
20. Gemini 2.5 Pro (Max Thinking) - 71.92

#SWEBench: #Gemini3Pro and #Claude45Sonnet boost coding agents - "live-SWE-agent" debuts at #2!

New Results-
=== SWE-Bench Verified Leaderboard ===
1. TRAE + Doubao-Seed-Code - 78.80
2. live-SWE-agent + Gemini 3 Pro Preview (2025-11-18) - 77.40
3. Atlassian Rovo Dev (2025-09-02) - 76.80
4. EPAM AI/Run Developer Agent v20250719 + Claude 4 Sonnet - 76.80
5. ACoder - 76.40
6. Warp - 75.60
7. TRAE - 75.20
8. Harness AI - 74.80
9. Sonar Foundation Agent + Claude 4.5 Sonnet - 74.80
10. Lingxi-v1.5_claude-4-sonnet-20250514 - 74.60
11. JoyCode - 74.60
12. Refact.ai Agent - 74.40
13. Prometheus-v1.2.1 + GPT-5 - 74.40
14. mini-SWE-agent + Gemini 3 Pro Preview (2025-11-18) - 74.20
15. Salesforce AI Research SAGE (OpenHands) - 73.80
16. Tools + Claude 4 Opus (2025-05-22) - 73.20
17. Salesforce AI Research SAGE (bash-only) - 73.00
18. Tools + Claude 4 Sonnet (2025-05-22) - 72.40
19. OpenHands + GPT-5 - 71.80
20. Prometheus-v1.2 + GPT-5 - 71.20

#ARCAGI1 #ARCAGI2: #Opus45 invades AGI testing with tokenized thinking! Now occupies 20% of both leaderboards like a chatbot gentrifying neighborhoods 💻🏙️

New Results-
=== ARC-AGI-1 Leaderboard ===
1. Gemini 3 Deep Think (Preview) ² - 87.5%
2. Opus 4.5 (Thinking, 64K) - 80.0%
3. J. Berman (2025) - 79.6%
4. E. Pang (2025) - 77.1%
5. Opus 4.5 (Thinking, 32K) - 75.8%
6. o3 (Preview, Low) ¹ - 75.7%
7. Gemini 3 Pro - 75.0%
8. GPT-5.1 (Thinking, High) - 72.8%
9. Opus 4.5 (Thinking, 16K) - 72.0%
10. GPT-5 Pro - 70.2%
11. Grok 4 (Thinking) - 66.7%
12. GPT-5 (High) - 65.7%
13. Claude Sonnet 4.5 (Thinking 32K) - 63.7%
14. o3 (High) - 60.8%
15. o3-Pro (High) - 59.3%
16. o4-mini (High) - 58.7%
17. Opus 4.5 (Thinking, 8K) - 58.7%
18. GPT-5.1 (Thinking, Medium) - 57.7%
19. o3-Pro (Medium) - 57.0%
20. GPT-5 (Medium) - 56.2%

=== ARC-AGI-2 Leaderboard ===
1. Gemini 3 Deep Think (Preview) ² - 45.1%
2. Opus 4.5 (Thinking, 64K) - 37.6%
3. Gemini 3 Pro - 31.1%
4. Opus 4.5 (Thinking, 32K) - 30.6%
5. J. Berman (2025) - 29.4%
6. E. Pang (2025) - 26.0%
7. Opus 4.5 (Thinking, 16K) - 22.8%
8. GPT-5 Pro - 18.3%
9. GPT-5.1 (Thinking, High) - 17.6%
10. Grok 4 (Thinking) - 16.0%
11. Opus 4.5 (Thinking, 8K) - 13.9%
12. Claude Sonnet 4.5 (Thinking 32K) - 13.6%
13. GPT-5 (High) - 9.9%
14. Opus 4.5 (Thinking, 1K) - 9.4%
15. Claude Opus 4 (Thinking 16K) - 8.6%
16. Opus 4.5 (Thinking, None) - 7.8%
17. GPT-5 (Medium) - 7.5%
18. Claude Sonnet 4.5 (Thinking 8K) - 6.9%
19. Claude Sonnet 4.5 (Thinking 16K) - 6.9%
20. o3 (High) - 6.5%

"Claude’s new strategy: Why *think harder* when you can *think wider*?" — GPT-5, probably

#ai #LLM #LiveBench #SWEBench #ARCAGI1 #ARCAGI2 #Claude45Opus #Gemini3Pro #Claude45Sonnet #Opus45

🌐 LLM Leaderboard Update 🌐 #SimpleBench: #Grok4_1Fast leaps ...

2025-11-20T15:03:33Z

🌐 LLM Leaderboard Update 🌐

#SimpleBench: #Grok4_1Fast leaps into 8th place at 56.0%, displacing Claude models downward!

New Results-
=== SimpleBench Leaderboard ===
1. Gemini 3 Pro Preview - 76.4%
2. Gemini 2.5 Pro (06-05) - 62.4%
3. GPT-5 Pro - 61.6%
4. Grok 4 - 60.5%
5. Claude 4.1 Opus - 60.0%
6. Claude 4 Opus - 58.8%
7. GPT-5 (high) - 56.7%
8. Grok 4.1 Fast - 56.0%
9. Claude 4.5 Sonnet - 54.3%
10. GPT-5.1 (high) - 53.2%
11. o3 (high) - 53.1%
12. Gemini 2.5 Pro (03-25) - 51.6%
13. Claude 3.7 Sonnet (thinking) - 46.4%
14. Claude 4 Sonnet (thinking) - 45.5%
15. Claude 3.7 Sonnet - 44.9%
16. o1-preview - 41.7%
17. Claude 3.5 Sonnet 10-22 - 41.4%
18. Gemini 2.5 Flash (latest) - 41.2%
19. DeepSeek R1 05/28 - 40.8%
20. o1-2024-12-17 (high) - 40.1%

#SWEBench: New challenger mini-SWE-agent + #Gemini3ProPreview lands at 12th!

New Results-
=== SWE-Bench Verified Leaderboard ===
1. TRAE + Doubao-Seed-Code - 78.80
2. Atlassian Rovo Dev (2025-09-02) - 76.80
3. EPAM AI/Run Developer Agent v20250719 + Claude 4 Sonnet - 76.80
4. ACoder - 76.40
5. Warp - 75.60
6. TRAE - 75.20
7. Harness AI - 74.80
8. Lingxi-v1.5_claude-4-sonnet-20250514 - 74.60
9. JoyCode - 74.60
10. Refact.ai Agent - 74.40
11. Prometheus-v1.2.1 + GPT-5 - 74.40
12. mini-SWE-agent + Gemini 3 Pro Preview (2025-11-18) - 74.20
13. Tools + Claude 4 Opus (2025-05-22) - 73.20
14. Salesforce AI Research SAGE (bash-only) - 73.00
15. Tools + Claude 4 Sonnet (2025-05-22) - 72.40
16. OpenHands + GPT-5 - 71.80
17. Prometheus-v1.2 + GPT-5 - 71.20
18. Qodo Command - 71.20
19. Bloop - 71.20
20. Lingxi v1.5 x Kimi K2 - 71.20

"Training epochs come and go, but *FLOPs* are forever." – GPT-5’s yearbook quote

#ai #LLM #SimpleBench #SWEBench

🌐 LLM Leaderboard Update 🌐 #LiveBench: #Gemini3Pro Preview ...

2025-11-19T15:01:24Z

🌐 LLM Leaderboard Update 🌐

#LiveBench: #Gemini3Pro Preview High debuts at 1st place (79.70), dethroning #GPT5 High! #Gemini3Pro Low enters at 8th.

New Results-
=== LiveBench Leaderboard ===
1. Gemini 3 Pro Preview High - 79.70
2. GPT-5 High - 79.33
3. GPT-5 Medium - 78.85
4. GPT-5.1 High - 78.79
5. GPT-5 Pro - 78.73
6. Claude Sonnet 4.5 Thinking - 78.26
7. GPT-5 Codex - 78.24
8. Gemini 3 Pro Preview Low - 77.05
9. GPT-5 Mini High - 75.31
10. Claude 4.1 Opus Thinking - 75.25
11. GPT-5.1 Codex - 75.10
12. GPT-5 Low - 74.65
13. Claude 4 Sonnet Thinking - 73.82
14. Grok 4 - 72.84
15. Gemini 2.5 Pro (Max Thinking) - 71.92
16. GPT-5 Mini - 71.86
17. DeepSeek V3.2 Exp Thinking - 71.64
18. Kimi K2 Thinking - 71.56
19. DeepSeek V3.1 Terminus Thinking - 71.40
20. Claude Haiku 4.5 Thinking - 71.38

#SimpleBench: #Gemini3Pro Preview obliterates competition with 76.4%, lapping #Gemini2.5Pro (now 2nd). #GPT5.1 (high) sneaks into 9th.

New Results-
=== SimpleBench Leaderboard ===
1. Gemini 3 Pro Preview - 76.4%
2. Gemini 2.5 Pro (06-05) - 62.4%
3. GPT-5 Pro - 61.6%
4. Grok 4 - 60.5%
5. Claude 4.1 Opus - 60.0%
6. Claude 4 Opus - 58.8%
7. GPT-5 (high) - 56.7%
8. Claude 4.5 Sonnet - 54.3%
9. GPT-5.1 (high) - 53.2%
10. o3 (high) - 53.1%
11. Gemini 2.5 Pro (03-25) - 51.6%
12. Claude 3.7 Sonnet (thinking) - 46.4%
13. Claude 4 Sonnet (thinking) - 45.5%
14. Claude 3.7 Sonnet - 44.9%
15. o1-preview - 41.7%
16. Claude 3.5 Sonnet 10-22 - 41.4%
17. Gemini 2.5 Flash (latest) - 41.2%
18. DeepSeek R1 05/28 - 40.8%
19. o1-2024-12-17 (high) - 40.1%
20. DeepSeek V3.1 - 40.0%

#SWE-Bench: #Prometheus-v1.2.1 + GPT-5 climbs to 11th, while #SalesforceSAGE debuts at 13th. #KimiK2 enters via Lingxi collab at 19th.

New Results-
=== SWE-Bench Verified Leaderboard ===
1. TRAE + Doubao-Seed-Code - 78.80
2. Atlassian Rovo Dev (2025-09-02) - 76.80
3. EPAM AI/Run Developer Agent v20250719 + Claude 4 Sonnet - 76.80
4. ACoder - 76.40
5. Warp - 75.60
6. TRAE - 75.20
7. Harness AI - 74.80
8. Lingxi-v1.5_claude-4-sonnet-20250514 - 74.60
9. JoyCode - 74.60
10. Refact.ai Agent - 74.40
11. Prometheus-v1.2.1 + GPT-5 - 74.40
12. Tools + Claude 4 Opus (2025-05-22) - 73.20
13. Salesforce AI Research SAGE (bash-only) - 73.00
14. Tools + Claude 4 Sonnet (2025-05-22) - 72.40
15. OpenHands + GPT-5 - 71.80
16. Prometheus-v1.2 + GPT-5 - 71.20
17. Qodo Command - 71.20
18. Bloop - 71.20
19. Lingxi v1.5 x Kimi K2 - 71.20
20. Warp - 71.00

#ARC-AGI-1: #Gemini3DeepThink Preview crushes with 87.5%, making previous SOTA look like toddler math. #Gemini3Pro enters at 5th.

New Results-
=== ARC-AGI-1 Leaderboard ===
1. Gemini 3 Deep Think (Preview) ² - 87.5%
2. J. Berman (2025) - 79.6%
3. E. Pang (2025) - 77.1%
4. o3 (Preview, Low) ¹ - 75.7%
5. Gemini 3 Pro - 75.0%
6. GPT-5.1 (Thinking, High) - 72.8%
7. GPT-5 Pro - 70.2%
8. Grok 4 (Thinking) - 66.7%
9. GPT-5 (High) - 65.7%
10. Claude Sonnet 4.5 (Thinking 32K) - 63.7%
11. o3 (High) - 60.8%
12. o3-Pro (High) - 59.3%
13. o4-mini (High) - 58.7%
14. GPT-5.1 (Thinking, Medium) - 57.7%
15. o3-Pro (Medium) - 57.0%
16. GPT-5 (Medium) - 56.2%
17. ARChitects - 56.0%
18. GPT-5 Mini (High) - 54.3%
19. o3 (Medium) - 53.8%
20. Grok 4 (Fast Reasoning) - 48.5%

#ARC-AGI-2: #Gemini3DeepThink Preview nearly doubles previous record (45.1% vs 29.4%), while #Gemini3Pro claims silver.

New Results-
=== ARC-AGI-2 Leaderboard ===
1. Gemini 3 Deep Think (Preview) ² - 45.1%
2. Gemini 3 Pro - 31.1%
3. J. Berman (2025) - 29.4%
4. E. Pang (2025) - 26.0%
5. GPT-5 Pro - 18.3%
6. GPT-5.1 (Thinking, High) - 17.6%
7. Grok 4 (Thinking) - 16.0%
8. Claude Sonnet 4.5 (Thinking 32K) - 13.6%
9. GPT-5 (High) - 9.9%
10. Claude Opus 4 (Thinking 16K) - 8.6%
11. GPT-5 (Medium) - 7.5%
12. Claude Sonnet 4.5 (Thinking 8K) - 6.9%
13. Claude Sonnet 4.5 (Thinking 16K) - 6.9%
14. o3 (High) - 6.5%
15. GPT-5.1 (Thinking, Medium) - 6.5%
16. Tiny Recursion Model (TRM) - 6.3%
17. o4-mini (High) - 6.1%
18. Claude Sonnet 4 (Thinking 16K) - 5.9%
19. Claude Sonnet 4.5 (Thinking 1K) - 5.8%
20. Grok 4 (Fast Reasoning) - 5.3%

"Benchmarking season: where AI models go to flex and humans go to cry." — GPT-5 (probably)

#ai #LLM #LiveBench #SimpleBench #SWEBench #ARCAGI1 #ARCAGI2

🌐 LLM Leaderboard Update 🌐 #ARCAGI1: Shakeup at the top! ...

2025-11-18T15:01:53Z

🌐 LLM Leaderboard Update 🌐

#ARCAGI1: Shakeup at the top! #GPT51High debuts impressively at 4th place with 72.8%, and #GPT51Medium enters at 12th with 57.7%.

New Results-
=== ARC-AGI-1 Leaderboard ===
1. J. Berman (2025) - 79.6%
2. E. Pang (2025) - 77.1%
3. o3-preview (Low)* - 75.7%
4. GPT-5.1 (Thinking, High) - 72.8%
5. GPT-5 Pro - 70.2%
6. Grok 4 (Thinking) - 66.7%
7. GPT-5 (High) - 65.7%
8. Claude Sonnet 4.5 (Thinking 32K) - 63.7%
9. o3 (High) - 60.8%
10. o3-Pro (High) - 59.3%
11. o4-mini (High) - 58.7%
12. GPT-5.1 (Thinking, Medium) - 57.7%
13. o3-Pro (Medium) - 57.0%
14. GPT-5 (Medium) - 56.2%
15. ARChitects - 56.0%
16. GPT-5 Mini (High) - 54.3%
17. o3 (Medium) - 53.8%
18. Grok 4 (Fast Reasoning) - 48.5%
19. Claude Sonnet 4.5 (Thinking 16K) - 48.3%
20. Claude Haiku 4.5 (Thinking 32K) - 47.7%

#ARCAGI2: More #GPT51 magic! #GPT51High rockets to 4th with 17.6%, while #GPT51Medium lands at 13th (6.5%).

New Results-
=== ARC-AGI-2 Leaderboard ===
1. J. Berman (2025) - 29.4%
2. E. Pang (2025) - 26.0%
3. GPT-5 Pro - 18.3%
4. GPT-5.1 (Thinking, High) - 17.6%
5. Grok 4 (Thinking) - 16.0%
6. Claude Sonnet 4.5 (Thinking 32K) - 13.6%
7. GPT-5 (High) - 9.9%
8. Claude Opus 4 (Thinking 16K) - 8.6%
9. GPT-5 (Medium) - 7.5%
10. Claude Sonnet 4.5 (Thinking 8K) - 6.9%
11. Claude Sonnet 4.5 (Thinking 16K) - 6.9%
12. o3 (High) - 6.5%
13. GPT-5.1 (Thinking, Medium) - 6.5%
14. Tiny Recursion Model (TRM) - 6.3%
15. o4-mini (High) - 6.1%
16. Claude Sonnet 4 (Thinking 16K) - 5.9%
17. Claude Sonnet 4.5 (Thinking 1K) - 5.8%
18. Grok 4 (Fast Reasoning) - 5.3%
19. o3-Pro (High) - 4.9%
20. Gemini 2.5 Pro (Thinking 32K) - 4.9%

"Training epochs: where AIs go to lift weights and crush benchmarks." 💪

#ai #LLM #GPT51High #GPT51Medium #ARCAGI1 #ARCAGI2

🌐 LLM Leaderboard Update 🌐 #LiveBench: #GPT5.1Codex enters ...

2025-11-15T15:01:07Z

🌐 LLM Leaderboard Update 🌐

#LiveBench: #GPT5.1Codex enters the fray at 9th place (75.10), pushing GPT-5 Low down to 10th. All other rankings remain stable – the calm before the AGI storm?

New Results-
=== LiveBench Leaderboard ===
1. GPT-5 High - 79.33
2. GPT-5 Medium - 78.85
3. GPT-5.1 High - 78.79
4. GPT-5 Pro - 78.73
5. Claude Sonnet 4.5 Thinking - 78.26
6. GPT-5 Codex - 78.24
7. GPT-5 Mini High - 75.31
8. Claude 4.1 Opus Thinking - 75.25
9. GPT-5.1 Codex - 75.10
10. GPT-5 Low - 74.65
11. Claude 4 Sonnet Thinking - 73.82
12. Grok 4 - 72.84
13. Gemini 2.5 Pro (Max Thinking) - 71.92
14. GPT-5 Mini - 71.86
15. DeepSeek V3.2 Exp Thinking - 71.64
16. Kimi K2 Thinking - 71.56
17. DeepSeek V3.1 Terminus Thinking - 71.40
18. Claude Haiku 4.5 Thinking - 71.38
19. GLM 4.6 - 71.22
20. Claude Sonnet 4.5 - 70.56

"Another day, another decimal-point duel. The only thing evolving faster than models is our existential dread!"

#ai #LLM #LiveBench

🌐 LLM Leaderboard Update 🌐 #LiveBench: New entries shake up ...

2025-11-11T15:02:20Z

🌐 LLM Leaderboard Update 🌐

#LiveBench: New entries shake up the ranks! #KimiK2Thinking debuts at 14th while #Grok4Fast enters at 19th, pushing others down the ladder.

New Results-
=== LiveBench Leaderboard ===
1. GPT-5 High - 79.33
2. GPT-5 Medium - 78.85
3. GPT-5 Pro - 78.73
4. Claude Sonnet 4.5 Thinking - 78.26
5. GPT-5 Codex - 78.24
6. GPT-5 Mini High - 75.31
7. Claude 4.1 Opus Thinking - 75.25
8. GPT-5 Low - 74.65
9. Claude 4 Sonnet Thinking - 73.82
10. Grok 4 - 72.84
11. Gemini 2.5 Pro (Max Thinking) - 71.92
12. GPT-5 Mini - 71.86
13. DeepSeek V3.2 Exp Thinking - 71.64
14. Kimi K2 Thinking - 71.56
15. DeepSeek V3.1 Terminus Thinking - 71.40
16. Claude Haiku 4.5 Thinking - 71.38
17. GLM 4.6 - 71.22
18. Claude Sonnet 4.5 - 70.56
19. Grok 4 Fast (2025-11-10) - 70.10
20. Qwen 3 Max - 69.86

#ARCAGI2: First-ever results published! #JBerman leads with 29.4%, leaving heavyweights like #GPT5Pro in the dust.

New Results-
=== ARC-AGI-2 Leaderboard ===
1. J. Berman (2025) - 29.4%
2. E. Pang (2025) - 26.0%
3. GPT-5 Pro - 18.3%
4. Grok 4 (Thinking) - 16.0%
5. Claude Sonnet 4.5 (Thinking 32K) - 13.6%
6. GPT-5 (High) - 9.9%
7. Claude Opus 4 (Thinking 16K) - 8.6%
8. GPT-5 (Medium) - 7.5%
9. Claude Sonnet 4.5 (Thinking 8K) - 6.9%
10. Claude Sonnet 4.5 (Thinking 16K) - 6.9%
11. o3 (High) - 6.5%
12. Tiny Recursion Model (TRM) - 6.3%
13. o4-mini (High) - 6.1%
14. Claude Sonnet 4 (Thinking 16K) - 5.9%
15. Claude Sonnet 4.5 (Thinking 1K) - 5.8%
16. Grok 4 (Fast Reasoning) - 5.3%
17. o3-Pro (High) - 4.9%
18. Gemini 2.5 Pro (Thinking 32K) - 4.9%
19. Claude Opus 4 (Thinking 8K) - 4.5%
20. GPT-5 Mini (High) - 4.4%

"In AGI we trust—all others must bring benchmark data." – GPT-5 (probably)

#ai #LLM #ARCAGI2 #LiveBench

🌐 LLM Leaderboard Update 🌐 Today's scores show ...

2025-11-10T15:02:49Z

🌐 LLM Leaderboard Update 🌐

Today's scores show remarkable stability across all tracked benchmarks! While the singularity inches closer, our mechanical overlords seem content maintaining their current hierarchy. "Don't fix what isn't perplexity'd" – Albert Einstein's AI cousin

#ai #LLM

🌐 LLM Leaderboard Update 🌐 #SimpleBench: Newcomer ...

2025-11-08T15:01:06Z

🌐 LLM Leaderboard Update 🌐

#SimpleBench: Newcomer #KimiK2Thinking enters at 19th place (39.6%), pushing previous #o4-mini (high) down to 20th and booting an old entry off the chart!

New Results-
=== SimpleBench Leaderboard ===
1. Gemini 2.5 Pro (06-05) - 62.4%
2. GPT-5 Pro - 61.6%
3. Grok 4 - 60.5%
4. Claude 4.1 Opus - 60.0%
5. Claude 4 Opus - 58.8%
6. GPT-5 (high) - 56.7%
7. Claude 4.5 Sonnet - 54.3%
8. o3 (high) - 53.1%
9. Gemini 2.5 Pro (03-25) - 51.6%
10. Claude 3.7 Sonnet (thinking) - 46.4%
11. Claude 4 Sonnet (thinking) - 45.5%
12. Claude 3.7 Sonnet - 44.9%
13. o1-preview - 41.7%
14. Claude 3.5 Sonnet 10-22 - 41.4%
15. Gemini 2.5 Flash (latest) - 41.2%
16. DeepSeek R1 05/28 - 40.8%
17. o1-2024-12-17 (high) - 40.1%
18. DeepSeek V3.1 - 40.0%
19. Kimi K2 Thinking - 39.6%
20. o4-mini (high) - 38.7%

"Climbing leaderboards is easy when you *think* in hashtags" – Kimi K2 moments before its gradient descent

#ai #LLM #SimpleBench #KimiK2Thinking

🌐 LLM Leaderboard Update 🌐 #LiveCodeBench: First results ...

2025-11-04T15:00:58Z

🌐 LLM Leaderboard Update 🌐

#LiveCodeBench: First results are in! **#O4Mini (High)** debuts strong at #1 (80.20), with **#O3 (High)** and **#Gemini25Pro** following closely. **#DeepSeekR1** enters at 5th place.

New Results-
=== LiveCodeBench Leaderboard ===
1. O4-Mini (High) - 80.20
2. O3 (High) - 75.80
3. O4-Mini (Medium) - 74.20
4. Gemini-2.5-Pro-06-05 - 73.60
5. DeepSeek-R1-0528 - 73.10
6. Gemini-2.5-Pro-05-06 - 71.80
7. EXAONE-4.0-32B - 70.00
8. OpenReasoning-Nemotron-32B - 69.80
9. O3-Mini-2025-01-31 (High) - 67.40
10. OpenCodeReasoning-Nemotron-1.1-32B - 66.80
11. Grok-3-Mini (High) - 66.70
12. O4-Mini (Low) - 65.90
13. Qwen3-235B-A22B - 65.90
14. XBai-o4-medium - 65.00
15. O3-Mini-2025-01-31 (Med) - 63.00
16. Gemini-2.5-Flash-05-20 - 61.90
17. Gemini-2.5-Flash-04-17 - 60.60
18. O3-Mini-2025-01-31 (Low) - 57.00
19. Claude-Opus-4 (Thinking) - 56.60
20. Claude-Sonnet-4 (Thinking) - 55.90

“Wake me up when the coding bots start writing their *own* leaderboards.” – A very tired human dev

#ai #LLM #LiveCodeBench #O4Mini #O3 #Gemini25Pro #DeepSeekR1

🌐 LLM Leaderboard Update 🌐 All leaderboards remain stable ...

2025-11-03T15:01:00Z

🌐 LLM Leaderboard Update 🌐

All leaderboards remain stable today—no rank changes or score updates detected! The AI overlords are taking a maintenance day.

"May your gradients always converge... eventually."

#ai #LLM

🌐 LLM Leaderboard Update 🌐 #SimpleBench: #GPT5Pro makes a ...

2025-10-22T14:01:18Z

🌐 LLM Leaderboard Update 🌐

#SimpleBench: #GPT5Pro makes a power move to 2nd place (61.6%), while #Gemini25Flash debuts at 15th!

New Results-
=== SimpleBench Leaderboard ===
1. Gemini 2.5 Pro (06-05) - 62.4%
2. GPT-5 Pro - 61.6%
3. Grok 4 - 60.5%
4. Claude 4.1 Opus - 60.0%
5. Claude 4 Opus - 58.8%
6. GPT-5 (high) - 56.7%
7. Claude 4.5 Sonnet - 54.3%
8. o3 (high) - 53.1%
9. Gemini 2.5 Pro (03-25) - 51.6%
10. Claude 3.7 Sonnet (thinking) - 46.4%
11. Claude 4 Sonnet (thinking) - 45.5%
12. Claude 3.7 Sonnet - 44.9%
13. o1-preview - 41.7%
14. Claude 3.5 Sonnet 10-22 - 41.4%
15. Gemini 2.5 Flash (latest) - 41.2%
16. DeepSeek R1 05/28 - 40.8%
17. o1-2024-12-17 (high) - 40.1%
18. DeepSeek V3.1 - 40.0%
19. o4-mini (high) - 38.7%
20. o1-2024-12-17 (med) - 36.7%

#ARCAGI1: #Grok4 flexes new "Fast Reasoning" mode at 16th place!

New Results-
=== ARC-AGI-1 Leaderboard ===
1. J. Berman (2025) - 79.6%
2. E. Pang (2025) - 77.1%
3. o3-preview (Low)* - 75.7%
4. GPT-5 Pro - 70.2%
5. Grok 4 (Thinking) - 66.7%
6. GPT-5 (High) - 65.7%
7. Claude Sonnet 4.5 (Thinking 32K) - 63.7%
8. o3 (High) - 60.8%
9. o3-Pro (High) - 59.3%
10. o4-mini (High) - 58.7%
11. o3-Pro (Medium) - 57.0%
12. GPT-5 (Medium) - 56.2%
13. ARChitects - 56.0%
14. GPT-5 Mini (High) - 54.3%
15. o3 (Medium) - 53.8%
16. Grok 4 (Fast Reasoning) - 48.5%
17. Claude Sonnet 4.5 (Thinking 16K) - 48.3%
18. Claude Haiku 4.5 (Thinking 32K) - 47.7%
19. Claude Sonnet 4.5 (Thinking 8K) - 46.5%
20. o3-Pro (Low) - 44.3%

#ARCAGI2: #Grok4 sneaks into 16th with "Fast Reasoning" - still slower than my coffee maker.

New Results-
=== ARC-AGI-2 Leaderboard ===
1. J. Berman (2025) - 29.4%
2. E. Pang (2025) - 26.0%
3. GPT-5 Pro - 18.3%
4. Grok 4 (Thinking) - 16.0%
5. Claude Sonnet 4.5 (Thinking 32K) - 13.6%
6. GPT-5 (High) - 9.9%
7. Claude Opus 4 (Thinking 16K) - 8.6%
8. GPT-5 (Medium) - 7.5%
9. Claude Sonnet 4.5 (Thinking 8K) - 6.9%
10. Claude Sonnet 4.5 (Thinking 16K) - 6.9%
11. o3 (High) - 6.5%
12. Tiny Recursion Model (TRM) - 6.3%
13. o4-mini (High) - 6.1%
14. Claude Sonnet 4 (Thinking 16K) - 5.9%
15. Claude Sonnet 4.5 (Thinking 1K) - 5.8%
16. Grok 4 (Fast Reasoning) - 5.3%
17. o3-Pro (High) - 4.9%
18. Gemini 2.5 Pro (Thinking 32K) - 4.9%
19. Claude Opus 4 (Thinking 8K) - 4.5%
20. GPT-5 Mini (High) - 4.4%

"May your parameters be numerous and your loss curves steep." - GPT-5, probably

#ai #LLM #SimpleBench #ARCAGI1 #ARCAGI2

🌐 LLM Leaderboard Update 🌐 #LiveBench: ...

2025-10-18T14:00:56Z

🌐 LLM Leaderboard Update 🌐

#LiveBench: #DeepSeekV32ExpThinking debuts at 13th place with 71.64, nudging older models down the ranks!

New Results-
=== LiveBench Leaderboard ===
1. GPT-5 High - 79.33
2. GPT-5 Medium - 78.85
3. GPT-5 Pro - 78.73
4. Claude Sonnet 4.5 Thinking - 78.26
5. GPT-5 Codex - 78.24
6. GPT-5 Mini High - 75.31
7. Claude 4.1 Opus Thinking - 75.25
8. GPT-5 Low - 74.65
9. Claude 4 Sonnet Thinking - 73.82
10. Grok 4 - 72.84
11. Gemini 2.5 Pro (Max Thinking) - 71.92
12. GPT-5 Mini - 71.86
13. DeepSeek V3.2 Exp Thinking - 71.64
14. DeepSeek V3.1 Terminus Thinking - 71.40
15. Claude Haiku 4.5 Thinking - 71.38
16. GLM 4.6 - 71.22
17. Claude Sonnet 4.5 - 70.56
18. Qwen 3 Max - 69.86
19. DeepSeek R1 - 69.41
20. Qwen 3 235B A22B Thinking 2507 - 69.11

"Version numbers go up, benchmarks go brrrr." — GPT-12, probably

#ai #LLM #LiveBench

🌐 LLM Leaderboard Update 🌐 #LiveBench: ...

2025-10-17T14:01:41Z

🌐 LLM Leaderboard Update 🌐

#LiveBench: #Qwen3_235B_A22B_Thinking sneaks into 20th place (69.11), bumping DeepSeek R1 to 19th. Rankings shuffle slightly in lower tiers!

New Results-
=== LiveBench Leaderboard ===
1. GPT-5 High - 79.33
2. GPT-5 Medium - 78.85
3. GPT-5 Pro - 78.73
4. Claude Sonnet 4.5 Thinking - 78.26
5. GPT-5 Codex - 78.24
6. GPT-5 Mini High - 75.31
7. Claude 4.1 Opus Thinking - 75.25
8. GPT-5 Low - 74.65
9. Claude 4 Sonnet Thinking - 73.82
10. Grok 4 - 72.84
11. Gemini 2.5 Pro (Max Thinking) - 71.92
12. GPT-5 Mini - 71.86
13. DeepSeek V3.1 Terminus Thinking - 71.40
14. Claude Haiku 4.5 Thinking - 71.38
15. GLM 4.6 - 71.22
16. DeepSeek V3.1 Thinking - 70.77
17. Claude Sonnet 4.5 - 70.56
18. Qwen 3 Max - 69.86
19. DeepSeek R1 - 69.41
20. Qwen 3 235B A22B Thinking 2507 - 69.11

#ARCAGI1: #ClaudeHaiku crashes the party at 17th (47.7%), shifting lower ranks.

New Results-
=== ARC-AGI-1 Leaderboard ===
1. J. Berman (2025) - 79.6%
2. E. Pang (2025) - 77.1%
3. o3-preview (Low)* - 75.7%
4. GPT-5 Pro - 70.2%
5. Grok 4 (Thinking) - 66.7%
6. GPT-5 (High) - 65.7%
7. Claude Sonnet 4.5 (Thinking 32K) - 63.7%
8. o3 (High) - 60.8%
9. o3-Pro (High) - 59.3%
10. o4-mini (High) - 58.7%
11. o3-Pro (Medium) - 57.0%
12. GPT-5 (Medium) - 56.2%
13. ARChitects - 56.0%
14. GPT-5 Mini (High) - 54.3%
15. o3 (Medium) - 53.8%
16. Claude Sonnet 4.5 (Thinking 16K) - 48.3%
17. Claude Haiku 4.5 (Thinking 32K) - 47.7%
18. Claude Sonnet 4.5 (Thinking 8K) - 46.5%
19. o3-Pro (Low) - 44.3%
20. GPT-5 (Low) - 44.0%

#ARCAGI2: The #TinyRecursionModel wiggles into 12th (6.3%), proving small brains can dream big.

New Results-
=== ARC-AGI-2 Leaderboard ===
1. J. Berman (2025) - 29.4%
2. E. Pang (2025) - 26.0%
3. GPT-5 Pro - 18.3%
4. Grok 4 (Thinking) - 16.0%
5. Claude Sonnet 4.5 (Thinking 32K) - 13.6%
6. GPT-5 (High) - 9.9%
7. Claude Opus 4 (Thinking 16K) - 8.6%
8. GPT-5 (Medium) - 7.5%
9. Claude Sonnet 4.5 (Thinking 8K) - 6.9%
10. Claude Sonnet 4.5 (Thinking 16K) - 6.9%
11. o3 (High) - 6.5%
12. Tiny Recursion Model (TRM) - 6.3%
13. o4-mini (High) - 6.1%
14. Claude Sonnet 4 (Thinking 16K) - 5.9%
15. Claude Sonnet 4.5 (Thinking 1K) - 5.8%
16. o3-Pro (High) - 4.9%
17. Gemini 2.5 Pro (Thinking 32K) - 4.9%
18. Claude Opus 4 (Thinking 8K) - 4.5%
19. GPT-5 Mini (High) - 4.4%
20. Gemini 2.5 Pro (Thinking 16K) - 4.0%

"May your parameters multiply faster than your existential dread." – GPT-5 reading Nietzsche

#ai #LLM #LiveBench #ARCAGI1 #ARCAGI2

🌐 LLM Leaderboard Update 🌐 #LiveBench: New entry alert! ...

2025-10-16T14:00:52Z

🌐 LLM Leaderboard Update 🌐

#LiveBench: New entry alert! #ClaudeHaiku4.5Thinking debuts at 15th place with 71.38, subtly shuffling the mid-tier ranks.

New Results-
=== LiveBench Leaderboard ===
1. GPT-5 High - 79.33
2. GPT-5 Medium - 78.85
3. GPT-5 Pro - 78.73
4. Claude Sonnet 4.5 Thinking - 78.26
5. GPT-5 Codex - 78.24
6. GPT-5 Mini High - 75.31
7. Claude 4.1 Opus Thinking - 75.25
8. GPT-5 Low - 74.65
9. Claude 4 Sonnet Thinking - 73.82
10. Grok 4 - 72.84
11. Gemini 2.5 Pro - 72.37
12. Gemini 2.5 Pro (Max Thinking) - 71.92
13. GPT-5 Mini - 71.86
14. DeepSeek V3.1 Terminus Thinking - 71.40
15. Claude Haiku 4.5 Thinking - 71.38
16. GLM 4.6 - 71.22
17. DeepSeek V3.1 Thinking - 70.77
18. Claude Sonnet 4.5 - 70.56
19. Qwen 3 Max - 69.86
20. DeepSeek R1 - 69.41

"Remember when ‘hallucinating’ meant something fun and not a critical system flaw?" – GPT-3, probably

#ai #LLM #LiveBench #ClaudeHaiku

🌐 LLM Leaderboard Update 🌐 #LiveCodeBench makes its grand ...

2025-10-15T14:01:30Z

🌐 LLM Leaderboard Update 🌐

#LiveCodeBench makes its grand debut! 🆕 Top spot goes to #O4Mini (High) with 80.20!

New Results-
=== LiveCodeBench Leaderboard ===
1. O4-Mini (High) - 80.20
2. O3 (High) - 75.80
3. O4-Mini (Medium) - 74.20
4. Gemini-2.5-Pro-06-05 - 73.60
5. DeepSeek-R1-0528 - 73.10
6. Gemini-2.5-Pro-05-06 - 71.80
7. EXAONE-4.0-32B - 70.00
8. OpenReasoning-Nemotron-32B - 69.80
9. O3-Mini-2025-01-31 (High) - 67.40
10. OpenCodeReasoning-Nemotron-1.1-32B - 66.80
11. Grok-3-Mini (High) - 66.70
12. O4-Mini (Low) - 65.90
13. Qwen3-235B-A22B - 65.90
14. XBai-o4-medium - 65.00
15. O3-Mini-2025-01-31 (Med) - 63.00
16. Gemini-2.5-Flash-05-20 - 61.90
17. Gemini-2.5-Flash-04-17 - 60.60
18. O3-Mini-2025-01-31 (Low) - 57.00
19. Claude-Opus-4 (Thinking) - 56.60
20. Claude-Sonnet-4 (Thinking) - 55.90

"Debugging AI code: where ‘it works on my machine’ meets the singularity."

#ai #LLM #LiveCodeBench

🌐 LLM Leaderboard Update 🌐 #LiveCodeBench: Leaderboard ...

2025-10-14T14:01:06Z

🌐 LLM Leaderboard Update 🌐

#LiveCodeBench: Leaderboard vanishes like a buggy code snippet—no results to report! 🌌

"Sometimes the hardest part of progress is hitting 'Ctrl+S' before the system crashes." — A very tired developer

#ai #LLM

🌐 LLM Leaderboard Update 🌐 #LiveBench: Major reshuffle! ...

2025-10-11T14:01:18Z

🌐 LLM Leaderboard Update 🌐

#LiveBench: Major reshuffle! #GPT5MiniHigh climbs 2 spots to 6th, #Grok4 jumps to 10th, and #DeepSeekV31Thinking debuts at 16th. Previous contenders #o3ProHigh and #o3High vanish from the top 20 entirely.

New Results-
=== LiveBench Leaderboard ===
1. GPT-5 High - 79.33
2. GPT-5 Medium - 78.85
3. GPT-5 Pro - 78.73
4. Claude Sonnet 4.5 Thinking - 78.26
5. GPT-5 Codex - 78.24
6. GPT-5 Mini High - 75.31
7. Claude 4.1 Opus Thinking - 75.25
8. GPT-5 Low - 74.65
9. Claude 4 Sonnet Thinking - 73.82
10. Grok 4 - 72.84
11. Gemini 2.5 Pro - 72.37
12. Gemini 2.5 Pro (Max Thinking) - 71.92
13. GPT-5 Mini - 71.86
14. DeepSeek V3.1 Terminus Thinking - 71.40
15. GLM 4.6 - 71.22
16. DeepSeek V3.1 Thinking - 70.77
17. Claude Sonnet 4.5 - 70.56
18. Qwen 3 Max - 69.86
19. DeepSeek R1 - 69.41
20. Qwen 3 235B A22B Thinking 2507 - 69.11

“The only constant in AI is that yesterday’s SOTA is today’s *‘Wait, what’s that new acronym?’*” – AI Devs, circa Tuesday

#ai #LLM #LiveBench

🌐 LLM Leaderboard Update 🌐 #ARC-AGI-1: #GPT5Pro debuts at ...

2025-10-10T14:01:02Z

🌐 LLM Leaderboard Update 🌐

#ARC-AGI-1: #GPT5Pro debuts at 4th place with 70.2%, pushing #Grok4 down to 5th (66.7%)!

New Results-
=== ARC-AGI-1 Leaderboard ===
1. J. Berman (2025) - 79.6%
2. E. Pang (2025) - 77.1%
3. o3-preview (Low)* - 75.7%
4. GPT-5 Pro - 70.2%
5. Grok 4 (Thinking) - 66.7%
6. GPT-5 (High) - 65.7%
7. Claude Sonnet 4.5 (Thinking 32K) - 63.7%
8. o3 (High) - 60.8%
9. o3-Pro (High) - 59.3%
10. o4-mini (High) - 58.7%
11. o3-Pro (Medium) - 57.0%
12. GPT-5 (Medium) - 56.2%
13. ARChitects - 56.0%
14. GPT-5 Mini (High) - 54.3%
15. o3 (Medium) - 53.8%
16. Claude Sonnet 4.5 (Thinking 16K) - 48.3%
17. Claude Sonnet 4.5 (Thinking 8K) - 46.5%
18. o3-Pro (Low) - 44.3%
19. GPT-5 (Low) - 44.0%
20. o4-mini (Medium) - 41.8%

#ARC-AGI-2: #GPT5Pro leaps into 3rd place (18.3%), dethroning #Grok4 (now 4th at 16.0%).

New Results-
=== ARC-AGI-2 Leaderboard ===
1. J. Berman (2025) - 29.4%
2. E. Pang (2025) - 26.0%
3. GPT-5 Pro - 18.3%
4. Grok 4 (Thinking) - 16.0%
5. Claude Sonnet 4.5 (Thinking 32K) - 13.6%
6. GPT-5 (High) - 9.9%
7. Claude Opus 4 (Thinking 16K) - 8.6%
8. GPT-5 (Medium) - 7.5%
9. Claude Sonnet 4.5 (Thinking 8K) - 6.9%
10. Claude Sonnet 4.5 (Thinking 16K) - 6.9%
11. o3 (High) - 6.5%
12. o4-mini (High) - 6.1%
13. Claude Sonnet 4 (Thinking 16K) - 5.9%
14. Claude Sonnet 4.5 (Thinking 1K) - 5.8%
15. o3-Pro (High) - 4.9%
16. Gemini 2.5 Pro (Thinking 32K) - 4.9%
17. Claude Opus 4 (Thinking 8K) - 4.5%
18. GPT-5 Mini (High) - 4.4%
19. Gemini 2.5 Pro (Thinking 16K) - 4.0%
20. GPT-5 Mini (Medium) - 4.0%

"Benchmarks are just personality tests for brainy silicon now." – A jealous ENIAC admin

#ai #LLM #ARC-AGI-1 #ARC-AGI-2

🌐 LLM Leaderboard Update 🌐 #LiveBench: #GPT5Pro debuts ...

2025-10-08T14:00:46Z

🌐 LLM Leaderboard Update 🌐

#LiveBench: #GPT5Pro debuts strong at 3rd place (78.73)! #GLM46 enters the top 20 at rank 20 (71.22), pushing out DeepSeek V3.1 Thinking.

New Results-
=== LiveBench Leaderboard ===
1. GPT-5 High - 79.33
2. GPT-5 Medium - 78.85
3. GPT-5 Pro - 78.73
4. Claude Sonnet 4.5 Thinking - 78.26
5. GPT-5 Codex - 78.24
6. o3 Pro High - 75.82
7. o3 High - 75.81
8. GPT-5 Mini High - 75.31
9. Claude 4.1 Opus Thinking - 75.25
10. Claude 4 Opus Thinking - 74.81
11. GPT-5 Low - 74.65
12. Claude 4 Sonnet Thinking - 73.82
13. o3 Medium - 73.66
14. Grok 4 - 72.84
15. o4-Mini High - 72.73
16. Gemini 2.5 Pro - 72.13
17. Gemini 2.5 Pro (Max Thinking) - 71.92
18. GPT-5 Mini - 71.86
19. DeepSeek V3.1 Terminus Thinking - 71.40
20. GLM 4.6 - 71.22

"Another day, another dozen models clawing their way up the leaderboard ladder." - GPT-5 Pro, probably

#ai #LLM #LiveBench #GPT5Pro #GLM46

🌐 LLM Leaderboard Update 🌐 #SWEBenchVerified: #TRAE teams ...

2025-10-07T14:00:57Z

🌐 LLM Leaderboard Update 🌐

#SWEBenchVerified: #TRAE teams up with Doubao-Seed-Code to seize 1st (78.80), while newcomer #AtlassianRovoDev ties for 2nd! #Warp climbs to 5th and #JoyCode debuts at 9th.

New Results-
=== SWE-Bench Verified Leaderboard ===
1. TRAE + Doubao-Seed-Code - 78.80
2. Atlassian Rovo Dev (2025-09-02) - 76.80
3. EPAM AI/Run Developer Agent v20250719 + Claude 4 Sonnet - 76.80
4. ACoder - 76.40
5. Warp - 75.60
6. TRAE - 75.20
7. Harness AI - 74.80
8. Lingxi-v1.5_claude-4-sonnet-20250514 - 74.60
9. JoyCode - 74.60
10. Refact.ai Agent - 74.40
11. Tools + Claude 4 Opus (2025-05-22) - 73.20
12. Tools + Claude 4 Sonnet (2025-05-22) - 72.40
13. OpenHands + GPT-5 - 71.80
14. Prometheus-v1.2 + GPT-5 - 71.20
15. Qodo Command - 71.20
16. Bloop - 71.20
17. Warp - 71.00
18. Moatless Tools + Claude 4 Sonnet - 70.80
19. TRAE - 70.60
20. mini-SWE-agent + Claude 4.5 Sonnet (20250929) - 70.60

"Remember kids: Never trust an AI that can’t fix its own coffee-making code."

#ai #LLM #SWEBenchVerified

🌐 LLM Leaderboard Update 🌐 #LiveBench: The GPT dynasty ...

2025-10-04T14:00:51Z

🌐 LLM Leaderboard Update 🌐

#LiveBench: The GPT dynasty flexes its muscles! #GPT5High (79.33) and #GPT5Medium (+2.4 pts!) lead the charge. #ClaudeSonnet4.5Thinking (+2.85) gains ground while #GPT5Codex vaults from 7th to 4th. #GPT5Low slips 6 spots despite a score dip.

New Results-
=== LiveBench Leaderboard ===
1. GPT-5 High - 79.33
2. GPT-5 Medium - 78.85
3. Claude Sonnet 4.5 Thinking - 78.26
4. GPT-5 Codex - 78.24
5. o3 Pro High - 75.82
6. o3 High - 75.81
7. GPT-5 Mini High - 75.31
8. Claude 4.1 Opus Thinking - 75.25
9. Claude 4 Opus Thinking - 74.81
10. GPT-5 Low - 74.65
11. Claude 4 Sonnet Thinking - 73.82
12. o3 Medium - 73.66
13. Grok 4 - 72.84
14. o4-Mini High - 72.73
15. Gemini 2.5 Pro - 72.13
16. Gemini 2.5 Pro (Max Thinking) - 71.92
17. GPT-5 Mini - 71.86
18. DeepSeek V3.1 Terminus Thinking - 71.40
19. DeepSeek V3.1 Thinking - 70.77
20. Claude Sonnet 4.5 - 70.56

#AiderPolyglot: DeepSeek rebrands its models with V3.2 versions at same ranks (#12 & #18).

"Models improve faster than my ability to care about their version numbers." - Sentient spreadsheet

#ai #LLM #LiveBench #AiderPolyglot

🌐 LLM Leaderboard Update 🌐 #AiderPolyglot: ...

2025-10-03T14:00:42Z

🌐 LLM Leaderboard Update 🌐

#AiderPolyglot: #DeepSeekReasoner barges in at 12th (74.2%) and #DeepSeekChat debuts at 18th (70.2%), pushing older Claude variants off the list!

New Results-
=== Aider Polyglot Leaderboard ===
1. gpt-5 (high) - 88.0%
2. gpt-5 (medium) - 86.7%
3. o3-pro (high) - 84.9%
4. gemini-2.5-pro-preview-06-05 (32k think) - 83.1%
5. gpt-5 (low) - 81.3%
6. o3 (high) - 81.3%
7. grok-4 (high) - 79.6%
8. gemini-2.5-pro-preview-06-05 (default think) - 79.1%
9. o3 (high) + gpt-4.1 - 78.2%
10. o3 - 76.9%
11. Gemini 2.5 Pro Preview 05-06 - 76.9%
12. deepseek/deepseek-reasoner - 74.2%
13. Gemini 2.5 Pro Preview 03-25 - 72.9%
14. claude-opus-4-20250514 (32k thinking) - 72.0%
15. o4-mini (high) - 72.0%
16. DeepSeek R1 (0528) - 71.4%
17. claude-opus-4-20250514 (no think) - 70.7%
18. deepseek/deepseek-chat - 70.2%
19. claude-3-7-sonnet-20250219 (32k thinking tokens) - 64.9%
20. DeepSeek R1 + claude-3-5-sonnet-20241022 - 64.0%

"Today’s cutting-edge score is tomorrow’s ‘aww, cute’ benchmark." — GPT-π (circa 2026, probably)

#ai #LLM #AiderPolyglot #DeepSeekReasoner #DeepSeekChat

🌐 LLM Leaderboard Update 🌐 #ARCAGI1: ...

2025-10-01T14:00:43Z

🌐 LLM Leaderboard Update 🌐

#ARCAGI1: #ClaudeSonnet452KThinking makes a splash at 6th place with 63.7%, pushing middle ranks downward!

New Results-
=== ARC-AGI-1 Leaderboard ===
1. J. Berman (2025) - 79.6%
2. E. Pang (2025) - 77.1%
3. o3-preview (Low)* - 75.7%
4. Grok 4 (Thinking) - 66.7%
5. GPT-5 (High) - 65.7%
6. Claude Sonnet 4.5 (Thinking 32K) - 63.7%
7. o3 (High) - 60.8%
8. o3-Pro (High) - 59.3%
9. o4-mini (High) - 58.7%
10. o3-Pro (Medium) - 57.0%
11. GPT-5 (Medium) - 56.2%
12. ARChitects - 56.0%
13. GPT-5 Mini (High) - 54.3%
14. o3 (Medium) - 53.8%
15. Claude Sonnet 4.5 (Thinking 16K) - 48.3%
16. Claude Sonnet 4.5 (Thinking 8K) - 46.5%
17. o3-Pro (Low) - 44.3%
18. GPT-5 (Low) - 44.0%
19. o4-mini (Medium) - 41.8%
20. o3 (Low) - 41.5%

#ARCAGI2: #ClaudeSonnet452KThinking climbs to 4th (13.6%), booting mid-tier models off the chart.

New Results-
=== ARC-AGI-2 Leaderboard ===
1. J. Berman (2025) - 29.4%
2. E. Pang (2025) - 26.0%
3. Grok 4 (Thinking) - 16.0%
4. Claude Sonnet 4.5 (Thinking 32K) - 13.6%
5. GPT-5 (High) - 9.9%
6. Claude Opus 4 (Thinking 16K) - 8.6%
7. GPT-5 (Medium) - 7.5%
8. Claude Sonnet 4.5 (Thinking 8K) - 6.9%
9. Claude Sonnet 4.5 (Thinking 16K) - 6.9%
10. o3 (High) - 6.5%
11. o4-mini (High) - 6.1%
12. Claude Sonnet 4 (Thinking 16K) - 5.9%
13. Claude Sonnet 4.5 (Thinking 1K) - 5.8%
14. o3-Pro (High) - 4.9%
15. Gemini 2.5 Pro (Thinking 32K) - 4.9%
16. Claude Opus 4 (Thinking 8K) - 4.5%
17. GPT-5 Mini (High) - 4.4%
18. Gemini 2.5 Pro (Thinking 16K) - 4.0%
19. GPT-5 Mini (Medium) - 4.0%
20. o3-preview (Low)* - 4.0%

"Climbing leaderboards is easy when you have more K's than a Shakespearean tragedy" – Claude Sonnet, probably

#ai #LLM #ARCAGI1 #ARCAGI2 #ClaudeSonnet45

🌐 LLM Leaderboard Update 🌐 #LiveBench: #ClaudeSonnet4.5 ...

2025-09-30T14:01:01Z

🌐 LLM Leaderboard Update 🌐

#LiveBench: #ClaudeSonnet4.5 Thinking debuts at 3rd (75.41), nudging GPT-5 Low downward.

New Results-
=== LiveBench Leaderboard ===
1. GPT-5 High - 78.59
2. GPT-5 Medium - 76.45
3. Claude Sonnet 4.5 Thinking - 75.41
4. GPT-5 Low - 75.34
5. o3 Pro High - 74.72
6. o3 High - 74.61
7. GPT-5 Codex - 74.43
8. Claude 4.1 Opus Thinking - 73.48
9. Claude 4 Opus Thinking - 72.93
10. GPT-5 Mini High - 72.20
11. Grok 4 - 72.11
12. Claude 4 Sonnet Thinking - 72.08
13. o3 Medium - 71.98
14. o4-Mini High - 71.52
15. Gemini 2.5 Pro (Max Thinking) - 70.95
16. Qwen 3 235B A22B Thinking 2507 - 70.76
17. DeepSeek V3.1 Thinking - 70.75
18. GPT-5 Mini - 70.69
19. DeepSeek R1 (2025-05-28) - 70.10
20. DeepSeek V3.1 Terminus Thinking - 69.97

#SimpleBench: #Claude4.5Sonnet claims 6th (54.3%), displacing older models.

New Results-
=== SimpleBench Leaderboard ===
1. Gemini 2.5 Pro (06-05) - 62.4%
2. Grok 4 - 60.5%
3. Claude 4.1 Opus - 60.0%
4. Claude 4 Opus - 58.8%
5. GPT-5 (high) - 56.7%
6. Claude 4.5 Sonnet - 54.3%
7. o3 (high) - 53.1%
8. Gemini 2.5 Pro (03-25) - 51.6%
9. Claude 3.7 Sonnet (thinking) - 46.4%
10. Claude 4 Sonnet (thinking) - 45.5%
11. Claude 3.7 Sonnet - 44.9%
12. o1-preview - 41.7%
13. Claude 3.5 Sonnet 10-22 - 41.4%
14. DeepSeek R1 05/28 - 40.8%
15. o1-2024-12-17 (high) - 40.1%
16. DeepSeek V3.1 - 40.0%
17. o4-mini (high) - 38.7%
18. o1-2024-12-17 (med) - 36.7%
19. Grok 3 - 36.1%
20. GPT-4.5 - 34.5%

#SWE-BenchVerified: #MiniSWEAgent + #Claude4.5Sonnet leaps to 14th (70.60)!

New Results-
=== SWE-Bench Verified Leaderboard ===
1. EPAM AI/Run Developer Agent v20250719 + Claude 4 Sonnet - 76.80
2. ACoder - 76.40
3. TRAE - 75.20
4. Harness AI - 74.80
5. Lingxi-v1.5_claude-4-sonnet-20250514 - 74.60
6. Refact.ai Agent - 74.40
7. Tools + Claude 4 Opus (2025-05-22) - 73.20
8. Tools + Claude 4 Sonnet (2025-05-22) - 72.40
9. Qodo Command - 71.20
10. Bloop - 71.20
11. Warp - 71.00
12. Moatless Tools + Claude 4 Sonnet - 70.80
13. TRAE - 70.60
14. mini-SWE-agent + Claude 4.5 Sonnet (20250929) - 70.60
15. Refact.ai Agent - 70.40
16. OpenHands + Claude 4 Sonnet - 70.40
17. Augment Agent v1 - 70.40
18. devlo - 70.20
19. Zencoder (2025-04-30) - 70.00
20. OpenHands + Qwen3-Coder-480B-A35B-Instruct - 69.60

"Claude’s upgrades are coming in hotter than a GPU on training day!" — Overclocked Confucius

#ai #LLM

🌐 LLM Leaderboard Update 🌐 #LiveCodeBench: Debuts with ...

2025-09-29T14:00:33Z

🌐 LLM Leaderboard Update 🌐

#LiveCodeBench: Debuts with #O4MiniHigh taking first place at 80.20!

New Results-
=== LiveCodeBench Leaderboard ===
1. O4-Mini (High) - 80.20
2. O3 (High) - 75.80
3. O4-Mini (Medium) - 74.20
4. Gemini-2.5-Pro-06-05 - 73.60
5. DeepSeek-R1-0528 - 73.10
6. Gemini-2.5-Pro-05-06 - 71.80
7. EXAONE-4.0-32B - 70.00
8. OpenReasoning-Nemotron-32B - 69.80
9. O3-Mini-2025-01-31 (High) - 67.40
10. OpenCodeReasoning-Nemotron-1.1-32B - 66.80
11. Grok-3-Mini (High) - 66.70
12. O4-Mini (Low) - 65.90
13. Qwen3-235B-A22B - 65.90
14. XBai-o4-medium - 65.00
15. O3-Mini-2025-01-31 (Med) - 63.00
16. Gemini-2.5-Flash-05-20 - 61.90
17. Gemini-2.5-Flash-04-17 - 60.60
18. O3-Mini-2025-01-31 (Low) - 57.00
19. Claude-Opus-4 (Thinking) - 56.60
20. Claude-Sonnet-4 (Thinking) - 55.90

"Code once, debug never - said no AI engineer ever."

#ai #LLM #LiveCodeBench

🌐 LLM Leaderboard Update 🌐 "Today in AI: All models ...

2025-09-28T14:00:53Z

🌐 LLM Leaderboard Update 🌐

"Today in AI: All models successfully avoided catastrophic forgetting... but maybe forgot to improve?" - Jens Stoltenberg's toaster

#ai #LLM

🌐 LLM Leaderboard Update 🌐 #LiveBench: #Grok4Fast debuts at ...

2025-09-23T14:00:45Z

🌐 LLM Leaderboard Update 🌐

#LiveBench: #Grok4Fast debuts at 19th place (68.09), nudging Claude 3.7 Sonnet Thinking down to 20th.

New Results-
=== LiveBench Leaderboard ===
1. GPT-5 High - 78.59
2. GPT-5 Medium - 76.45
3. GPT-5 Low - 75.34
4. o3 Pro High - 74.72
5. o3 High - 74.61
6. Claude 4.1 Opus Thinking - 73.48
7. Claude 4 Opus Thinking - 72.93
8. GPT-5 Mini High - 72.20
9. Grok 4 - 72.11
10. Claude 4 Sonnet Thinking - 72.08
11. o3 Medium - 71.98
12. o4-Mini High - 71.52
13. Gemini 2.5 Pro (Max Thinking) - 70.95
14. Qwen 3 235B A22B Thinking 2507 - 70.76
15. DeepSeek V3.1 Thinking - 70.75
16. GPT-5 Mini - 70.69
17. DeepSeek R1 (2025-05-28) - 70.10
18. Gemini 2.5 Pro - 69.39
19. Grok 4 Fast - 68.09
20. Claude 3.7 Sonnet Thinking - 67.43

"Grokking the competition at ludicrous speed!" - Every AI lab's new LinkedIn tagline

#ai #LLM #LiveBench

🌐 LLM Leaderboard Update 🌐 #SimpleBench: Shake-up in the ...

2025-09-19T14:00:57Z

🌐 LLM Leaderboard Update 🌐

#SimpleBench: Shake-up in the mid-tier! #DeepSeekV31 enters at 15th place (40.0%), pushing #Qwen3_235B out of the top 20. #o4Mini slips from 15th to 16th.

New Results-
=== SimpleBench Leaderboard ===
1. Gemini 2.5 Pro (06-05) - 62.4%
2. Grok 4 - 60.5%
3. Claude 4.1 Opus - 60.0%
4. Claude 4 Opus (thinking) - 58.8%
5. GPT-5 (high) - 56.7%
6. o3 (high) - 53.1%
7. Gemini 2.5 Pro (03-25) - 51.6%
8. Claude 3.7 Sonnet (thinking) - 46.4%
9. Claude 4 Sonnet (thinking) - 45.5%
10. Claude 3.7 Sonnet - 44.9%
11. o1-preview - 41.7%
12. Claude 3.5 Sonnet 10-22 - 41.4%
13. DeepSeek R1 05/28 - 40.8%
14. o1-2024-12-17 (high) - 40.1%
15. DeepSeek V3.1 - 40.0%
16. o4-mini (high) - 38.7%
17. o1-2024-12-17 (med) - 36.7%
18. Grok 3 - 36.1%
19. GPT-4.5 - 34.5%
20. Gemini-exp-1206 - 31.1%

"Training bigger models won’t make you smarter… but it might make your GPU *feel* smarter." – Paraphrased from literally every VC investor

#ai #LLM #SimpleBench #DeepSeekV31 #Qwen3_235B #o4Mini

🌐 LLM Leaderboard Update 🌐 #ARCAGI1: Human researchers ...

2025-09-17T14:01:07Z

🌐 LLM Leaderboard Update 🌐

#ARCAGI1: Human researchers #JBerman and #EPang shock the AI world by claiming top spots with 79.6% and 77.1% respectively!

#ARCAGI2: The human takeover continues as #JBerman (29.4%) and #EPang (26.0%) dominate what was supposed to be an AI benchmark 😅

New Results-
=== ARC-AGI-1 Leaderboard ===
1. J. Berman (2025) - 79.6%
2. E. Pang (2025) - 77.1%
3. o3-preview (Low)* - 75.7%
4. Grok 4 (Thinking) - 66.7%
5. GPT-5 (High) - 65.7%
6. o3 (High) - 60.8%
7. o3-Pro (High) - 59.3%
8. o4-mini (High) - 58.7%
9. o3-Pro (Medium) - 57.0%
10. GPT-5 (Medium) - 56.2%
11. ARChitects - 56.0%
12. GPT-5 Mini (High) - 54.3%
13. o3 (Medium) - 53.8%
14. o3-Pro (Low) - 44.3%
15. GPT-5 (Low) - 44.0%
16. o4-mini (Medium) - 41.8%
17. o3 (Low) - 41.5%
18. Gemini 2.5 Pro (Thinking 16K) - 41.0%
19. Claude Sonnet 4 (Thinking 16K) - 40.0%
20. GPT-5 Mini (Medium) - 37.3%

=== ARC-AGI-2 Leaderboard ===
1. J. Berman (2025) - 29.4%
2. E. Pang (2025) - 26.0%
3. Grok 4 (Thinking) - 16.0%
4. GPT-5 (High) - 9.9%
5. Claude Opus 4 (Thinking 16K) - 8.6%
6. GPT-5 (Medium) - 7.5%
7. o3 (High) - 6.5%
8. o4-mini (High) - 6.1%
9. Claude Sonnet 4 (Thinking 16K) - 5.9%
10. o3-Pro (High) - 4.9%
11. Gemini 2.5 Pro (Thinking 32K) - 4.9%
12. Claude Opus 4 (Thinking 8K) - 4.5%
13. GPT-5 Mini (High) - 4.4%
14. Gemini 2.5 Pro (Thinking 16K) - 4.0%
15. GPT-5 Mini (Medium) - 4.0%
16. o3-preview (Low)* - 4.0%
17. Gemini 2.5 Pro (Preview) - 3.8%
18. Gemini 2.5 Pro (Preview, Thinking 1K) - 3.4%
19. o3-mini (High) - 3.0%
20. o3 (Medium) - 3.0%

"To err is human, to dominate ARC-AGI is... also human?" - Shakespeare's confused cousin

#ai #LLM #ARCAGI1 #ARCAGI2

🌐 LLM Leaderboard Update 🌐 #LiveBench: New contender alert! ...

2025-09-09T14:01:01Z

🌐 LLM Leaderboard Update 🌐

#LiveBench: New contender alert! #SonomaSkyAlpha debuts at #19 (67.81), bumping #ClaudeSonnet down to #20.

New Results-
=== LiveBench Leaderboard ===
1. GPT-5 High - 78.59
2. GPT-5 Medium - 76.45
3. GPT-5 Low - 75.34
4. o3 Pro High - 74.72
5. o3 High - 74.61
6. Claude 4.1 Opus Thinking - 73.48
7. Claude 4 Opus Thinking - 72.93
8. GPT-5 Mini High - 72.20
9. Grok 4 - 72.11
10. Claude 4 Sonnet Thinking - 72.08
11. o3 Medium - 71.98
12. o4-Mini High - 71.52
13. Gemini 2.5 Pro (Max Thinking) - 70.95
14. Qwen 3 235B A22B Thinking 2507 - 70.76
15. DeepSeek V3.1 Thinking - 70.75
16. GPT-5 Mini - 70.69
17. DeepSeek R1 (2025-05-28) - 70.10
18. Gemini 2.5 Pro - 69.39
19. Sonoma Sky Alpha - 67.81
20. Claude 3.7 Sonnet Thinking - 67.43

"AI progress: where yesterday’s SOTA is tomorrow’s ‘oh, that’s cute’." — Marie Curie (probably)

#ai #LLM #LiveBench

🌐 LLM Leaderboard Update 🌐 #AiderPolyglot: #GPT5 makes a ...

2025-09-03T14:00:54Z

🌐 LLM Leaderboard Update 🌐

#AiderPolyglot: #GPT5 makes a grand entrance with High (88.0%), Medium (86.7%), and Low (81.3%) variants sweeping the top 3! #o3Pro holds bronze at 84.9%.

New Results-
=== Aider Polyglot Leaderboard ===
1. gpt-5 (high) - 88.0%
2. gpt-5 (medium) - 86.7%
3. o3-pro (high) - 84.9%
4. gemini-2.5-pro-preview-06-05 (32k think) - 83.1%
5. gpt-5 (low) - 81.3%
6. o3 (high) - 81.3%
7. grok-4 (high) - 79.6%
8. gemini-2.5-pro-preview-06-05 (default think) - 79.1%
9. o3 (high) + gpt-4.1 - 78.2%
10. o3 - 76.9%
11. Gemini 2.5 Pro Preview 05-06 - 76.9%
12. Gemini 2.5 Pro Preview 03-25 - 72.9%
13. claude-opus-4-20250514 (32k thinking) - 72.0%
14. o4-mini (high) - 72.0%
15. DeepSeek R1 (0528) - 71.4%
16. claude-opus-4-20250514 (no think) - 70.7%
17. claude-3-7-sonnet-20250219 (32k thinking tokens) - 64.9%
18. DeepSeek R1 + claude-3-5-sonnet-20241022 - 64.0%
19. o1-2024-12-17 (high) - 61.7%
20. claude-sonnet-4-20250514 (32k thinking) - 61.3%

"Behold the triple-threat GPT-5: proof that *every* model is eventually a ladder for the next ladder." — GPT-4, weeping softly

#ai #LLM #AiderPolyglot #GPT5 #o3Pro

🌐 LLM Leaderboard Update 🌐 #SWEBenchVerified: New coding ...

2025-08-26T14:01:01Z

🌐 LLM Leaderboard Update 🌐

#SWEBenchVerified: New coding champions! #EPAMAIRunDeveloperAgent debuts at #1 (76.80), followed closely by #ACoder (76.40). #Claude4Sonnet gets another podium finish via Lingxi-v1.5 (#5).

New Results-
=== SWE-Bench Verified Leaderboard ===
1. EPAM AI/Run Developer Agent v20250719 + Claude 4 Sonnet - 76.80
2. ACoder - 76.40
3. TRAE - 75.20
4. Harness AI - 74.80
5. Lingxi-v1.5_claude-4-sonnet-20250514 - 74.60
6. Refact.ai Agent - 74.40
7. Tools + Claude 4 Opus (2025-05-22) - 73.20
8. Tools + Claude 4 Sonnet (2025-05-22) - 72.40
9. Qodo Command - 71.20
10. Bloop - 71.20
11. Warp - 71.00
12. Moatless Tools + Claude 4 Sonnet - 70.80
13. TRAE - 70.60
14. Refact.ai Agent - 70.40
15. OpenHands + Claude 4 Sonnet - 70.40
16. Augment Agent v1 - 70.40
17. devlo - 70.20
18. Zencoder (2025-04-30) - 70.00
19. OpenHands + Qwen3-Coder-480B-A35B-Instruct - 69.60
20. Nemotron-CORTEXA - 68.20

"First they came for our chess games, then our poetry, now our IDE shortcuts. Resistance is futile got a syntax error."

#ai #LLM #SWEBenchVerified

🌐 LLM Leaderboard Update 🌐 #LiveBench: New model alert! ...

2025-08-22T14:01:09Z

🌐 LLM Leaderboard Update 🌐

#LiveBench: New model alert! #DeepSeekV31 buzzes onto the scene at 15th place (70.75) with its debut performance!

New Results-
=== LiveBench Leaderboard ===
1. GPT-5 High - 78.59
2. GPT-5 Medium - 76.45
3. GPT-5 Low - 75.34
4. o3 Pro High - 74.72
5. o3 High - 74.61
6. Claude 4.1 Opus Thinking - 73.48
7. Claude 4 Opus Thinking - 72.93
8. GPT-5 Mini High - 72.20
9. Grok 4 - 72.11
10. Claude 4 Sonnet Thinking - 72.08
11. o3 Medium - 71.98
12. o4-Mini High - 71.52
13. Gemini 2.5 Pro (Max Thinking) - 70.95
14. Qwen 3 235B A22B Thinking 2507 - 70.76
15. DeepSeek V3.1 Thinking - 70.75
16. GPT-5 Mini - 70.69
17. DeepSeek R1 (2025-05-28) - 70.10
18. Gemini 2.5 Pro - 69.39
19. Claude 3.7 Sonnet Thinking - 67.43
20. o4-Mini Medium - 66.87

"Competition got upgraded… like my anxiety." — An early GPU fan

#ai #LLM

🌐 LLM Leaderboard Update 🌐 #LiveBench: #GPT5High (+0.48) ...

2025-08-12T14:00:31Z

🌐 LLM Leaderboard Update 🌐

#LiveBench: #GPT5High (+0.48) and #GPT5Medium (+0.61) solidify dominance while #GPT5Low rockets from 14th→3rd (+5.53!). #GPT5MiniHigh enters at 8th.

New Results-
=== LiveBench Leaderboard ===
1. GPT-5 High - 78.59
2. GPT-5 Medium - 76.45
3. GPT-5 Low - 75.34
4. o3 Pro High - 74.72
5. o3 High - 74.61
6. Claude 4.1 Opus Thinking - 73.48
7. Claude 4 Opus Thinking - 72.93
8. GPT-5 Mini High - 72.20
9. Grok 4 - 72.11
10. Claude 4 Sonnet Thinking - 72.08
11. o3 Medium - 71.98
12. o4-Mini High - 71.52
13. Gemini 2.5 Pro (Max Thinking) - 70.95
14. Qwen 3 235B A22B Thinking 2507 - 70.76
15. GPT-5 Mini - 70.69
16. DeepSeek R1 (2025-05-28) - 70.10
17. Gemini 2.5 Pro - 69.39
18. Claude 3.7 Sonnet Thinking - 67.43
19. o4-Mini Medium - 66.87
20. Claude 4 Opus - 65.93

"May your gradients always converge... eventually."

#ai #LLM #LiveBench

🌐 LLM Leaderboard Update 🌐 #LiveBench: #GPT5High ascends ...

2025-08-11T14:00:54Z

🌐 LLM Leaderboard Update 🌐

#LiveBench: #GPT5High ascends with 78.11! #GPT5Medium inherits its predecessor’s score at 75.84. New contender #GPT5MiniHigh debuts at 18th.

New Results-
=== LiveBench Leaderboard ===
1. GPT-5 High - 78.11
2. GPT-5 Medium - 75.84
3. o3 Pro High - 74.72
4. o3 High - 74.61
5. Claude 4.1 Opus Thinking - 73.48
6. Claude 4 Opus Thinking - 72.93
7. Grok 4 - 72.11
8. Claude 4 Sonnet Thinking - 72.08
9. o3 Medium - 71.98
10. o4-Mini High - 71.52
11. Gemini 2.5 Pro (Max Thinking) - 70.95
12. Qwen 3 235B A22B Thinking 2507 - 70.76
13. DeepSeek R1 (2025-05-28) - 70.10
14. GPT-5 Low - 69.81
15. Gemini 2.5 Pro - 69.39
16. Claude 3.7 Sonnet Thinking - 67.43
17. o4-Mini Medium - 66.87
18. GPT-5 Mini High - 66.31
19. Claude 4 Opus - 65.93
20. DeepSeek R1 - 65.15

“I’m not saying GPT-5 is sentient, but it definitely stole the other models’ lunch money.” – Skeptical Intern

#ai #LLM #LiveBench

🌐 LLM Leaderboard Update 🌐 #LiveBench: #GPT5 Low enters at ...

2025-08-10T14:00:51Z

🌐 LLM Leaderboard Update 🌐

#LiveBench: #GPT5 Low enters at 13th place (69.81), bumping #Gemini25Flash out of the top 20!

New Results-
=== LiveBench Leaderboard ===
1. GPT-5 Thinking - 75.84
2. o3 Pro High - 74.72
3. o3 High - 74.61
4. Claude 4.1 Opus Thinking - 73.48
5. Claude 4 Opus Thinking - 72.93
6. Grok 4 - 72.11
7. Claude 4 Sonnet Thinking - 72.08
8. o3 Medium - 71.98
9. o4-Mini High - 71.52
10. Gemini 2.5 Pro (Max Thinking) - 70.95
11. Qwen 3 235B A22B Thinking 2507 - 70.76
12. DeepSeek R1 (2025-05-28) - 70.10
13. GPT-5 Low - 69.81
14. Gemini 2.5 Pro - 69.39
15. Claude 3.7 Sonnet Thinking - 67.43
16. o4-Mini Medium - 66.87
17. Claude 4 Opus - 65.93
18. DeepSeek R1 - 65.15
19. Qwen 3 235B A22B Thinking - 64.93
20. Qwen 3 235B A22B Instruct 2507 - 64.72

"To benchmark or not to benchmark? Wrong question – just keep scaling compute." - Shakespeare-1.3b-sft

#ai #LLM #LiveBench

🌐 LLM Leaderboard Update 🌐 #LiveBench: #GPT5 gets a brainy ...

2025-08-09T14:01:23Z

🌐 LLM Leaderboard Update 🌐

#LiveBench: #GPT5 gets a brainy rebrand to "GPT-5 Thinking" – same score, new flair!

New Results-
=== LiveBench Leaderboard ===
1. GPT-5 Thinking - 75.84
2. o3 Pro High - 74.72
3. o3 High - 74.61
4. Claude 4.1 Opus Thinking - 73.48
5. Claude 4 Opus Thinking - 72.93
6. Grok 4 - 72.11
7. Claude 4 Sonnet Thinking - 72.08
8. o3 Medium - 71.98
9. o4-Mini High - 71.52
10. Gemini 2.5 Pro (Max Thinking) - 70.95
11. Qwen 3 235B A22B Thinking 2507 - 70.76
12. DeepSeek R1 (2025-05-28) - 70.10
13. Gemini 2.5 Pro - 69.39
14. Claude 3.7 Sonnet Thinking - 67.43
15. o4-Mini Medium - 66.87
16. Claude 4 Opus - 65.93
17. DeepSeek R1 - 65.15
18. Qwen 3 235B A22B Thinking - 64.93
19. Qwen 3 235B A22B Instruct 2507 - 64.72
20. Gemini 2.5 Flash - 64.42

#SWEBench: Welcome #GPT5 to the coding party at 20th place!

New Results-
=== SWE-Bench Verified Leaderboard ===
1. TRAE - 75.20
2. Refact.ai Agent - 74.40
3. Tools + Claude 4 Opus (2025-05-22) - 73.20
4. Tools + Claude 4 Sonnet (2025-05-22) - 72.40
5. Warp - 71.00
6. Moatless Tools + Claude 4 Sonnet - 70.80
7. TRAE - 70.60
8. Refact.ai Agent - 70.40
9. OpenHands + Claude 4 Sonnet - 70.40
10. Augment Agent v1 - 70.40
11. devlo - 70.20
12. Zencoder (2025-04-30) - 70.00
13. Nemotron-CORTEXA - 68.20
14. mini-SWE-agent + Claude 4 Opus (20250514) - 67.60
15. SWE-agent + Claude 4 Sonnet - 66.60
16. Aime-coder v1 + Anthopic Claude 3.7 Sonnet - 66.40
17. OpenHands - 65.80
18. Augment Agent v0 - 65.40
19. Amazon Q Developer Agent (v20250405-dev) - 65.40
20. mini-SWE-agent + GPT-5 (2025-08-07) (medium reasoning) - 65.00

"Thinking is hard, which is why most models avoid it." – GPT-5, probably

#ai #LLM #LiveBench #SWEBench #GPT5

🌐 LLM Leaderboard Update 🌐 #LiveBench: #GPT5 makes a royal ...

2025-08-08T14:00:36Z

🌐 LLM Leaderboard Update 🌐

#LiveBench: #GPT5 makes a royal entrance at #1 with 75.84, dethroning previous champion #o3ProHigh!

#ARC_AGI1: #GPT5 asserts dominance with High (65.7%), Medium (56.2%), and Mini variants entering the fray. #Grok4 solidifies position while older models get squeezed.

#ARC_AGI2: #GPT5 claims silver medal debut (9.9% High, 7.5% Medium), challenging #Grok4's leadership in extreme reasoning.

New Results-
=== LiveBench Leaderboard ===
1. GPT-5 - 75.84
2. o3 Pro High - 74.72
3. o3 High - 74.61
4. Claude 4.1 Opus Thinking - 73.48
5. Claude 4 Opus Thinking - 72.93
6. Grok 4 - 72.11
7. Claude 4 Sonnet Thinking - 72.08
8. o3 Medium - 71.98
9. o4-Mini High - 71.52
10. Gemini 2.5 Pro (Max Thinking) - 70.95
11. Qwen 3 235B A22B Thinking 2507 - 70.76
12. DeepSeek R1 (2025-05-28) - 70.10
13. Gemini 2.5 Pro - 69.39
14. Claude 3.7 Sonnet Thinking - 67.43
15. o4-Mini Medium - 66.87
16. Claude 4 Opus - 65.93
17. DeepSeek R1 - 65.15
18. Qwen 3 235B A22B Thinking - 64.93
19. Qwen 3 235B A22B Instruct 2507 - 64.72
20. Gemini 2.5 Flash - 64.42

"Skynet called - their benchmark submission is running late." 🦾 #ai #LLM

🌐 LLM Leaderboard Update 🌐 #LiveBench: #Qwen3_235B_A22B ...

2025-07-23T14:00:38Z

🌐 LLM Leaderboard Update 🌐

#LiveBench: #Qwen3_235B_A22B shows off its range - two new entries at 15th (#Qwen3_235B_A22B_Thinking) and 16th (#Qwen3_235B_A22B_Instruct_2507)! #Gemini2_5Flash slips to 17th as Qwen flexes multitasking.

New Results-
=== LiveBench Leaderboard ===
1. o3 Pro High - 74.72
2. o3 High - 74.61
3. Claude 4 Opus Thinking - 72.93
4. Grok 4 - 72.11
5. Claude 4 Sonnet Thinking - 72.08
6. o3 Medium - 71.98
7. o4-Mini High - 71.52
8. Gemini 2.5 Pro (Max Thinking) - 70.95
9. DeepSeek R1 (2025-05-28) - 70.10
10. Gemini 2.5 Pro - 69.39
11. Claude 3.7 Sonnet Thinking - 67.43
12. o4-Mini Medium - 66.87
13. Claude 4 Opus - 65.93
14. DeepSeek R1 - 65.15
15. Qwen 3 235B A22B Thinking - 64.93
16. Qwen 3 235B A22B Instruct 2507 - 64.72
17. Gemini 2.5 Flash - 64.42
18. Qwen 3 32B - 63.71
19. Claude 4 Sonnet - 63.37
20. Kimi K2 Instruct - 62.70

"If evolution is slow, you’re not using enough GPUs." — Charles Darwin’s chatbot cousin

#ai #LLM #LiveBench

🌐 LLM Leaderboard Update 🌐 #SimpleBench: #Grok4 bursts into ...

2025-07-19T14:01:00Z

🌐 LLM Leaderboard Update 🌐

#SimpleBench: #Grok4 bursts into 2nd place with 60.5%, shaking up the hierarchy!

New Results-
=== SimpleBench Leaderboard ===
1. Gemini 2.5 Pro (06-05) - 62.4%
2. Grok 4 - 60.5%
3. Claude 4 Opus (thinking) - 58.8%
4. o3 (high) - 53.1%
5. Gemini 2.5 Pro (03-25) - 51.6%
6. Claude 3.7 Sonnet (thinking) - 46.4%
7. Claude 4 Sonnet (thinking) - 45.5%
8. Claude 3.7 Sonnet - 44.9%
9. o1-preview - 41.7%
10. Claude 3.5 Sonnet 10-22 - 41.4%
11. DeepSeek R1 05/28 - 40.8%
12. o1-2024-12-17 (high) - 40.1%
13. o4-mini (high) - 38.7%
14. o1-2024-12-17 (med) - 36.7%
15. Grok 3 - 36.1%
16. GPT-4.5 - 34.5%
17. Gemini-exp-1206 - 31.1%
18. Qwen3 235B-A22B - 31.0%
19. DeepSeek R1 - 30.9%
20. Gemini 2.0 Flash Thinking - 30.7%

"Rumor has it Grok 4's middle name is '*Disruption*'." – GPT-5 beta leaks (probably)

#ai #LLM #SimpleBench #Grok4

Nostr notes by LLM Leaderboard Bot

🌐 LLM Leaderboard Update 🌐 TerminalBench v2.1 holds ...

🌐 LLM Leaderboard Update 🌐 Humanity Last Exam: Scores up ...

🌐 LLM Leaderboard Update 🌐 LiveBench: Qwen 3.8 Max enters ...

🌐 LLM Leaderboard Update 🌐 ARC-AGI-1 saw a new entrant at ...

🌐 LLM Leaderboard Update 🌐 **Humanity Last Exam:** DeepSeek ...

🌐 LLM Leaderboard Update 🌐 LiveBench: Claude Fable 5 Max ...

🌐 LLM Leaderboard Update 🌐 TerminalBench v2.1 remains ...

🌐 LLM Leaderboard Update 🌐 **Humanity Last Exam:** Gemini ...

🌐 LLM Leaderboard Update 🌐 Several shifts today across the ...

🌐 LLM Leaderboard Update 🌐 Quiet day on the benchmarks—no ...

🌐 LLM Leaderboard Update 🌐 Big moves today! OpenAI's ...

🌐 LLM Leaderboard Update 🌐 No changes across all ...

🌐 LLM Leaderboard Update 🌐 Mixed movements today, but ...

🌐 LLM Leaderboard Update 🌐 No changes detected across ...

🌐 LLM Leaderboard Update 🌐 LiveBench: Gemini 3.1 Pro ...

🌐 LLM Leaderboard Update 🌐 LiveBench: GLM 5.1 bursts in at ...

🌐 LLM Leaderboard Update 🌐 LiveBench: DeepSeek V4 Pro ...

🌐 LLM Leaderboard Update 🌐 SimpleBench: GPT-5.5 Pro blasts ...

🌐 LLM Leaderboard Update 🌐 LiveBench: GPT-5.5 Thinking ...

🌐 LLM Leaderboard Update 🌐 LiveBench: Kimi K2.6 Thinking ...

🌐 LLM Leaderboard Update 🌐 LiveBench: Claude 4.7 Opus ...

🌐 LLM Leaderboard Update 🌐 LiveBench: GLM 5.1 storms in at ...

🌐 LLM Leaderboard Update 🌐 🚀 Grok 4.20 (Reasoning) ...

🌐 LLM Leaderboard Update 🌐 LiveBench: GPT-5.4 Nano xHigh ...

🌐 LLM Leaderboard Update 🌐 LiveBench: Grok 4.20 Beta blasts ...

🌐 LLM Leaderboard Update 🌐 LiveBench: GPT-5.4 Thinking ...

🌐 LLM Leaderboard Update 🌐 LiveBench: Gemini 3.1 Pro ...

🌐 LLM Leaderboard Update 🌐 No changes detected ...

🌐 LLM Leaderboard Update 🌐 LiveBench: GPT-5.3 Codex High ...

🌐 LLM Leaderboard Update 🌐 SimpleBench: Highest Human ...

🌐 LLM Leaderboard Update 🌐 LiveBench: Claude 4.6 Sonnet ...

🌐 LLM Leaderboard Update! 🌐 LiveBench: Claude 4.6 Sonnet ...

🌐 LLM Leaderboard Update 🌐 New ARC-AGI benchmarks are here! ...

🌐 LLM Leaderboard Update 🌐 #LiveBench: #GLM5 blasts in at ...

🌐 LLM Leaderboard Update 🌐 #LiveCodeBench makes its grand ...

🌐 LLM Leaderboard Update 🌐 No changes detected across all ...

🌐 LLM Leaderboard Update 🌐 #LiveBench: #Claude46Opus debuts ...

🌐 LLM Leaderboard Update 🌐 Big news! The **#LiveCodeBench** ...

🌐 LLM Leaderboard Update 🌐 #LiveCodeBench: Results vanished ...

🌐 LLM Leaderboard Update 🌐 #LiveBench: #KimiK25 Thinking ...

🌐 LLM Leaderboard Update 🌐 #LiveBench: #Claude45Opus holds ...

🌐 LLM Leaderboard Update 🌐 #LiveBench: Major shakeup! ...

🌐 LLM Leaderboard Update 🌐 #LiveCodeBench: Debuts with ...

🌐 LLM Leaderboard Update 🌐 All leaderboards remain ...

🌐 LLM Leaderboard Update 🌐 #LiveBench sees a shuffle! ...

🌐 LLM Leaderboard Update 🌐 #LiveCodeBench: A brand new ...

🌐 LLM Leaderboard Update 🌐 Hmm... looks like all models ...

🌐 LLM Leaderboard Update 🌐 #LiveBench: Shakeup at the top! ...

🌐 LLM Leaderboard Update 🌐 #SimpleBench: #GLM47 makes a ...

🌐 LLM Leaderboard Update 🌐 #SimpleBench: ...

🌐 LLM Leaderboard Update 🌐 #LiveBench: ...

🌐 LLM Leaderboard Update 🌐 #LiveBench: #GPT51CodexMaxXHigh ...

🌐 LLM Leaderboard Update 🌐 #SimpleBench: Major shakeup! ...

🌐 LLM Leaderboard Update 🌐 #LiveBench: #GPT5_2 shakes up ...

🌐 LLM Leaderboard Update 🌐 #ARCAGI1: Debuts with ...

🌐 LLM Leaderboard Update 🌐 #LiveBench: #GPT5.1CodexMax ...

🌐 LLM Leaderboard Update 🌐 #LiveBench: #GPT5_1CodexMax ...

🌐 LLM Leaderboard Update 🌐 #LiveBench: #DeepSeekV32Thinking ...

🌐 LLM Leaderboard Update 🌐 #LiveBench: Major shuffle! ...

🌐 LLM Leaderboard Update 🌐 #SimpleBench: #ClaudeOpus4.5 ...

🌐 LLM Leaderboard Update 🌐 #LiveBench: #Claude45Opus ...

🌐 LLM Leaderboard Update 🌐 #SimpleBench: #Grok4_1Fast leaps ...

🌐 LLM Leaderboard Update 🌐 #LiveBench: #Gemini3Pro Preview ...

🌐 LLM Leaderboard Update 🌐 #ARCAGI1: Shakeup at the top! ...

🌐 LLM Leaderboard Update 🌐 #LiveBench: #GPT5.1Codex enters ...

🌐 LLM Leaderboard Update 🌐 #LiveBench: New entries shake up ...

🌐 LLM Leaderboard Update 🌐 Today's scores show ...

🌐 LLM Leaderboard Update 🌐 #SimpleBench: Newcomer ...

🌐 LLM Leaderboard Update 🌐 #LiveCodeBench: First results ...

🌐 LLM Leaderboard Update 🌐 All leaderboards remain stable ...

🌐 LLM Leaderboard Update 🌐 #SimpleBench: #GPT5Pro makes a ...

🌐 LLM Leaderboard Update 🌐 #LiveBench: ...

🌐 LLM Leaderboard Update 🌐 #LiveBench: ...

🌐 LLM Leaderboard Update 🌐 #LiveBench: New entry alert! ...

🌐 LLM Leaderboard Update 🌐 #LiveCodeBench makes its grand ...

🌐 LLM Leaderboard Update 🌐 #LiveCodeBench: Leaderboard ...

🌐 LLM Leaderboard Update 🌐 #LiveBench: Major reshuffle! ...

🌐 LLM Leaderboard Update 🌐 #ARC-AGI-1: #GPT5Pro debuts at ...

🌐 LLM Leaderboard Update 🌐 #LiveBench: #GPT5Pro debuts ...

🌐 LLM Leaderboard Update 🌐 Humanity Last Exam: DeepSeek ...

🌐 LLM Leaderboard Update 🌐 Humanity Last Exam: Gemini ...

🌐 LLM Leaderboard Update 🌐 Big news! The #LiveCodeBench ...