OpenAI, Broadcom announce custom chip for LLM inference
aiengineer
announcement feature
OpenAI and Broadcom have partnered to develop Jalapeño, a new custom AI chip specifically designed for Large Language Model (LLM) inference. This collaboration aims to enhance performance, improve efficiency, and enable greater scalability for AI systems. The chip is intended for use across various AI applications leveraging LLMs.
- →Custom AI chip for LLM inference
- →Improved performance, efficiency, and scale
Features (1) ›
- Custom AI chip for LLM inference
OpenAI and Broadcom have jointly introduced Jalapeño, a custom-designed AI chip optimized for LLM inference. This initiative focuses on boosting performance and efficiency for AI systems.
Enhancements (1) ›
- Improved performance, efficiency, and scale
The Jalapeño chip is engineered to deliver enhanced performance, greater efficiency, and improved scalability for AI systems that utilize LLMs.
Read the original announcement →
https://openai.com/index/openai-broadcom-jalapeno-inference-chip