OpenAI and Broadcom unveil LLM-optimized inference chip
Bùi Đăng Minh•Thứ ba, 30/6/2026, 09:00 (GMT+7)•4 min read
OpenAI and Broadcom introduce Jalapeño, a custom AI chip built for LLM inference to improve performance, efficiency, and scale across AI systems.
Nguồn / Original source: OpenAI (@sama & @OpenAIDevs)