OpenAI的Jalapeño芯片专为大规模快速推理而设计，基准测试显示其性能卓越。

OpenAI’s Jalapeño chip is built for fast inference at scale, benchmarks show | TechCrunch

🚨 Flash Sale 🚨 Get $100 off your Disrupt 2026 ticket

Back by popular demand: Save up to $300 on Disrupt

OpenAI’s Jalapeño chip is built for fast inference at scale, benchmarks show

At the Hot Chips conference on Tuesday, OpenAI shared a more detailed look at Jalapeño, including the first batch of benchmark results for the new system. Tested on SemiAnalysis’ InferenceX benchmark, Jalapeño registered both more tokens per user and more throughput per kilowatt than the currently available state-of-the-art inference processors.

“The bottom line is that the results show a very, very significant performance advance over state of the art,” said Richard Ho, OpenAI’s head of hardware, in a press call. “Jalapeño can serve more AI work per unit of power, while also returning responses more quickly. Its very efficient to serve a lot of customers, but it can also be very low latency.”

Notably, that comparison is against an Nvidia Blackwell system — but by the time Jalapeño reaches full deployment, the competition may have advanced significantly. Ho estimated that Jalapeño would deploy at the end of 2026 “in very small volumes,” with more significant deployment coming in 2027.

First announced last October, Jalapeño was developed by OpenAI in close collaboration with Broadcom, with OpenAIs own models assisting in the development process. The company plans to make Jalapeño a multigenerational platform, allowing AI products, models, chips, and memory all developed in concert.

Because of that full-stack approach, OpenAI was able to address specific phases in the inference process that often cause friction during inference processing. In particular, Jalapeño is designed to minimize delays during the prefill and communication phases of processing, which OpenAI says often act as bottlenecks.

“We designed Jalapeño to minimize data movement and communication delays,” the company said in a blog post presenting the results. “This means that model state, including the KV cache used while generating a response, can be explicitly placed and kept local while the system activates the right combination of compute, memory, and networking for each inference phase.”

When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.

Russell Brandom has been covering the tech industry since 2012, with a focus on platform policy and emerging technologies. He previously worked at The Verge and Rest of World, and has written for Wired, The Awl and MITs Technology Review.

He can be reached at russell.brandom@techcrunch.com or on Signal at 412-401-5489.

In less than 48 hours, your chance to save up to $300 on your tickets will end!

Two years after launch, Walmarts Flipkart is closing in on Indias quick-commerce leaders

Inherent, founded by DeepMind alumni, says its AI teammate just outperformed Anthropic and OpenAI at replicating research

Michael Polansky is training an AI model on skin that’s still alive

How AI accounting startup Rillet raised $100M and became a unicorn in 48 hours

Tesla’s solar roof is dead — here’s what went wrong

Oura faces lawsuit accusing it of misleading consumers about sleep-tracking accuracy

Home batteries are suddenly cheap and everywhere. Here’s why.