stevibe (@stevibe)새로 나온 Qwen3.6 35B-A3B를 여러 GPU(RTX 3090/4090/5090, DGX Spark)에서 돌려 성능을 비교한 벤치마크다. Ollama를 백엔드로 사용했으며, 지연시간(TTFT)과 토큰 처리 속도를 공개해 실사용 관점의 성능을 보여준다.https://x.com/stevibe/status/2045087373516492954#qwen #ollama #benchmark #llm #gpus
Related
I tried Siri AI, and so far it actually worksSiri, are you there? Parents want one thing, and one thing only, out of AI:...
I tried Siri AI, and so far it actually worksSiri, are you there? Parents want one thing, and one thing only, out of AI: to add a list of soccer games or "spirit week" theme days f...
OpenAI、IPOを非公開で申請 「リークを予想し自ら発表」 – CNET Japan https://www.yayafa.com/2819047/ #AgenticAi #AI #ArtificialGeneralIntelligen...
OpenAI、IPOを非公開で申請 「リークを予想し自ら発表」 – CNET Japan https://www.yayafa.com/2819047/ #AgenticAi #AI #ArtificialGeneralIntelligence #ArtificialIntelligence #ipo #OpenAI #エージェント型AI #人工知能 #汎用...
“The lawyers on both sides of a federal court case in #Mississippi were caught using artificial intelligence, a situatio...
“The lawyers on both sides of a federal court case in #Mississippi were caught using artificial intelligence, a situation where, effectively, generative #AI tools were used to argu...