started · updated
OpenAI unveils Jalapeno custom AI chip to rival Nvidia
OpenAI has unveiled performance data for its first custom AI inference chip, named Jalapeno, developed in collaboration with Broadcom and Celestica. Designed specifically to run large language models rather than train them, the chip aims to reduce operational costs and reliance on Nvidia hardware.
Benchmark testing conducted using SemiAnalysis' InferenceX tool indicates that Jalapeno outperforms Nvidia’s GB200 and GB300 systems in several key metrics. Specifically, the chip reportedly delivers between 1.5 and 1.9 times more AI work per watt and achieves 1.7 to 3.6 times lower latency. For highly interactive workloads, such as AI agents, performance advantages can reach up to 4.1 times.
OpenAI utilized its own advanced AI models to assist in the chip's design and circuit verification, which helped shorten the development cycle from concept to tapeout to just nine months. The chip is rated at 700 watts, though testing showed actual power draw remained at or below 550 watts. OpenAI expects to begin limited deployment of the Jalapeno chip by the end of 2026, with broader rollout following in 2027.
Entities
Claims
What the coverage asserts, and how many sources carry each claim.
- [○ 1 SOURCE] OpenAI plans to start deploying Jalapeno by the end of 2026. www.liputan6.com
- [● 3 SOURCES] Jalapeno produced 1.5 to 1.9 times more AI work per watt at peak throughput compared to Nvidia systems. www.ictbusiness.biz · www.soydemac.com · www.theindianpanorama.news
- [● 2 SOURCES] The Jalapeno chip is optimized for inference rather than model training. www.techbang.com · www.tekedia.com
- [● 3 SOURCES] Jalapeno demonstrated latency cuts between 1.7 and 3.6 times compared to Nvidia GB200 and GB300 systems. www.ictbusiness.biz · www.soydemac.com · www.theindianpanorama.news
- [● 5 SOURCES] OpenAI's custom Jalapeno chip is designed specifically for AI inference tasks. www.ictbusiness.biz · donanimgunlugu.com · www.techbang.com · www.soydemac.com · www.theindianpanorama.news
- [● 2 SOURCES] The design-to-tapeout process for the chip took nine months. www.ictbusiness.biz · donanimgunlugu.com
- [● 3 SOURCES] OpenAI used its own AI models to assist in the design of the Jalapeno chip. www.ictbusiness.biz · donanimgunlugu.com · www.techbang.com
- [● 3 SOURCES] The Jalapeno chip was developed in collaboration with Broadcom and Celestica. www.techbang.com · www.soydemac.com · www.theindianpanorama.news
- [● 4 SOURCES] Jalapeno produced between 1.5 and 1.9 times more AI work per watt at peak throughput compared to Nvidia systems. www.ictbusiness.biz · www.soydemac.com · www.theindianpanorama.news · www.tekedia.com
- [● 2 SOURCES] The Jalapeno chip is rated at 700 watts, though measured power draw during testing stayed at or below 550 watts. www.ictbusiness.biz · donanimgunlugu.com
- [● 3 SOURCES] Jalapeno demonstrated latency reductions between 1.7 and 3.6 times compared to Nvidia GB200 and GB300 systems. www.ictbusiness.biz · www.soydemac.com · www.theindianpanorama.news
- [○ 1 SOURCE] OpenAI plans to begin deploying the Jalapeno chip by the end of 2026. www.liputan6.com