API
< Back to all clusters
[TECHNOLOGY] · United States · 62 sources

started · updated

OpenAI unveils Jalapeño, custom AI inference chip

OpenAI has unveiled Jalapeño, its first custom AI inference chip developed in collaboration with Broadcom. Designed specifically for large language model (LLM) workloads, the chip aims to reduce OpenAI’s reliance on Nvidia hardware by providing more efficient and faster processing for AI inference tasks.

According to benchmark results from SemiAnalysis’ InferenceX, Jalapeño demonstrated significant performance advantages over existing systems like Nvidia’s GB200 and GB300. The chip reportedly delivers 1.5 to 1.9 times more AI work per watt at peak throughput and reduces end-to-end latency by 1.7 to 3.6 times. For highly interactive workloads, such as AI agents, performance gains reached up to 4.1 times.

OpenAI plans to begin small-scale deployment of the chip within its own computing infrastructure by the end of 2026, with a full production ramp expected between 2027 and 2028. The company noted that Jalapeño is the first in a multi-generational silicon platform, with second and third generations already in development. While the chip focuses on inference, OpenAI intends to continue using Nvidia and other partners' hardware for model training.

Entities

Broadcom · Nvidia · OpenAI · Richard Ho · Sam Altman · SemiAnalysis · TSMC

Claims

What the coverage asserts, and how many sources carry each claim.

Sources