OpenAI 称其首款自研推理芯片 Jalapeño 在速度和能效上超过 Nvidia GB200 与 GB300 系统。据 InferenceX 测试，在 GPT-OSS 120B、DeepSeek R1 670B 和 Kimi K2.5 1T 上，该芯片峰值吞吐下每瓦 AI 算力提升 1.5–1.9×，端到端延迟降低 1.7–3.6×，高交互负载性能提升 2.1–4.1×。
OpenAI 自研推理芯片的对比测试数据与 2026 年底部署计划一并给出，可据此了解其硬件自研路线的进展。
Holy: OpenAI says its first custom inference chip is already beating Nvidia GB200 and GB300 systems on speed and efficiency.
Across GPT‑OSS 120B, DeepSeek R1 670B, and Kimi K2.5 1T, Jalapeño delivered 1.5–1.9× more AI work per watt at peak throughput and 1.7–3.6× lower end-to-end latency in OpenAI’s own InferenceX testing!
For highly interactive workloads, OpenAI reports 2.1–4.1× higher performance.
The chip is rated at 700 watts, but remained at or below 550 watts during the tested workloads.
OpenAI plans to begin deploying Jalapeño by the end of 2026. Gen 2 is already deep in development, with Gen 3 taking shape.
Probably thats why Tibo said that in 1-2 years 750token/s will be the default
来源：@kimmonismus · x.com