
OpenAI's Jalapeño AI chip claims efficiency lead over Nvidia GB300 in first public benchmarks
Published by AINave Editorial • Reviewed by Ramit
OpenAI has released the first public benchmark results for its custom AI inference chip, Jalapeño, claiming it outperforms Nvidia's GB300 on tokens per watt and throughput per kilowatt. The results, presented at the Hot Chips conference at Stanford, are significant for AI builders because inference cost and power consumption are major constraints at scale. But the benchmarks are OpenAI's own, and deployment is still over a year away.
Jalapeño's performance claims
OpenAI's Jalapeño chip, developed with Broadcom and manufactured by TSMC, is designed specifically for inference workloads. Richard Ho, OpenAI's head of hardware, said the chip delivers greater AI output per watt while also cutting user response times. The benchmarks used SemiAnalysis' InferenceX test and ran models including OpenAI's smaller open-source model, DeepSeek, and Moonshot AI's Kimi model. The largest performance gap was on Kimi, the biggest model tested.
OpenAI normalized results to each accelerator's published TDP. Jalapeño is rated at 700 watts, but OpenAI said measured sustained power stayed at or below 550 watts. In contrast, Nvidia's GB300 is rated at 1,200 watts or more. On throughput per kilowatt, OpenAI reported [up to 1.9x better performance](https://247wallst.com/investing/2026/08/26/openai-says-its-new-chip-outperforms-nvidias-blackwell-as-nvidia-prepares-earnings-release
Sources
- OpenAI says its custom AI chip is beating Nvidia's best in benchmark tests
- OpenAI claims its new chip outperforms NVIDIA's Blackwell
- OpenAI says its 1st custom AI chip surpasses Nvidia systems in key...
- OpenAI Claims Its New Chips Can Outperform Nvidia... - Bloomberg
- OpenAI’s 700W Jalapeño ASIC outpaces 1,400W Nvidia flagship GPU...
- OpenAI's Jalapeño Chip Beats Nvidia in First… · AGI Hunt
- OpenAI Says Its New Chip Outperforms Nvidia's Blackwell As Nvidia...
- Meet Jalapeño: OpenAI’s First Custom AI Chip, Built With Broadcom
- ‘Superfast’: OpenAI’s first chip outpaces Nvidia systems with 1.9x more efficiency
- Nvidia once pledged up to $100B to OpenAI. Now OpenAI has built a chip designed to need less of Nvidia's hardware. Should investors worry?
- Nvidia Throws Its Support Behind OpenAI, Open AI, and… Super-Closed AI?
- Nvidia's AI moat is shifting from chips to capital
- Jim Chanos Says Nvidia Is Effectively Financing Its Own AI Chip Sales in Reported OpenAI Deal
- OpenAI Jalapeño chip beats Nvidia GB300 in benchmark tests
- OpenAI's chip beats Nvidia's best
- Nvidia once pledged up to $100B to OpenAI. Now OpenAI has built a chip designed to need less of Nvidia's hardware. Should investors worry?






















