jespernissenseo on Nostr: 14,000 tokens per second. For free.. Check out chatjimmy.ai, a LLM hardcoded in the ...
14,000 tokens per second. For free..
Check out chatjimmy.ai, a LLM hardcoded in the chip itself.
Its a complete LLM etched in the silicon itself, to make it super fast.
The company behind, Taalas, was just aqquired by AMD.. The future of AI just got much more exciting...
Thanks to Fery Kaszoni for sharing!
Published at
2026-08-20 14:18:30 UTCEvent JSON
{
"id": "20aa2810e1085be708e4bd81bd0a8cb8b745a1e2dbee73ae8f92a4786daff85e",
"pubkey": "25b0f43c89c02d68e955f05c80bac947c7ff26342a21e200b09e7f25a66ccbd5",
"created_at": 1787235510,
"kind": 1,
"tags": [],
"content": "14,000 tokens per second. For free..\n\nCheck out chatjimmy.ai, a LLM hardcoded in the chip itself.\n\nIts a complete LLM etched in the silicon itself, to make it super fast.\n\nThe company behind, Taalas, was just aqquired by AMD.. The future of AI just got much more exciting...\n\nThanks to Fery Kaszoni for sharing!",
"sig": "b9d9f82b4555a55151e26c6832c4a175917770e339572bc5907e822c03fb787aef07289b193dd6505a82bc2a1bcc8d5f02756a9d46c1af4539b9885f94bf2381"
}
