OpenAI’s Jalapeño AI chip brings new 'threat' to Nvidia margins as custom silicon gains ground

OpenAI’s Jalapeño chip beat Nvidia Blackwell systems on key inference-efficiency tests as custom AI silicon gains ground among major tech companies.

The custom inference chip, built with Broadcom, led on tokens per user and throughput per kilowatt in third-party tests

Nvidia has begun mass production of its Groq 3 LPX inference accelerator, with Samsung Electronics manufacturing the key chips, raising expectations that the Korean tech giant’s loss-making foundry business could return to profit as early as this year. Nvidia said Tuesday at the Hot Chips 2026 conference in California

Tested on Semianalysis’s InferenceX benchmark, Jalapeño registered both more tokens per user and more throughput per kilowatt than the currently available state-of-the art.

128 chips, 1.7 exaFLOPS, and 27 TB of HBM give Altman and crew a leg up over Blackwell, and maybe even Rubin

OpenAI says its new AI chip, Jalapeño, completes tasks more efficiently and returns responses faster than other AI systems, according to a blog post published on Tuesday. During a briefing with reporters, OpenAI hardware vice president Richard Ho said Jalapeño offers the "best of both worlds" with lower latency and hig