Jalapeño

Jalapeño is a custom ASIC for LLM inference, developed by OpenAI in partnership with Broadcom. It delivers approximately 50% lower inference costs and was accelerated using OpenAI’s own models, with deployment on Celestica rack technology.

Two-hop subgraph: this entity, every entity it directly relates to, and every entity those neighbors relate to. Drag a node, scroll to zoom, click to inspect — or click any neighbor and re-center the atlas there.