OpenAI and Broadcom unveil LLM-optimized inference chip
OpenAI and Broadcom announced a custom LLM inference chip to improve AI system performance, efficiency, and deployment scale.
Excerpt
OpenAI and Broadcom introduce Jalapeño, a custom AI chip built for LLM inference to improve performance, efficiency, and scale across AI systems.
Read at source: https://openai.com/index/openai-broadcom-jalapeno-inference-chip