ChipsAugust 25, 2026via The Verge AI
OpenAI says its Jalapeño chip can power faster AI responses than the competition
Why it matters
OpenAI is shipping custom silicon optimized for inference efficiency—a direct challenge to Nvidia's inference dominance and a signal that the frontier labs are building their own hardware stack to reduce dependency and control costs.
Key signals
- Jalapeño is an ASIC co-designed with Broadcom, first introduced June 2026
- Chip targets AI inference workloads specifically
- OpenAI claims lower latency AND higher throughput—typically a tradeoff
- Richard Ho (OpenAI VP Hardware) briefed reporters on performance claims
- Inference efficiency directly affects deployment economics and response speed for agents and real-time applications
The hook
OpenAI's Jalapeño chip breaks the latency-throughput tradeoff. Here's what that means for inference costs.
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 l…