OpenAI plans custom Jalapeño inference chip with Broadcom
Tech firms are pursuing custom AI chips to reduce reliance on Nvidia, TechCrunch reports.
Why it matters
The move reflects growing efforts by major AI and technology companies to reduce single-supplier risk in AI compute. Custom inference chips could reshape how leading firms manage cost, supply and control over AI infrastructure.
The key points
- 1.OpenAI is planning Jalapeño, a custom inference chip with Broadcom.
- 2.Google, Apple and SpaceX are also building their own chips.
- 3.The trend challenges total dependence on Nvidia AI chips.
OpenAI has shared plans for Jalapeño, a custom inference chip built with Broadcom, according to TechCrunch. The report says OpenAI joins Google, Apple and SpaceX among companies developing their own chips as Nvidia continues to dominate the AI chip market.
Sources & original reporting
This brief summarizes and links to reporting from the publishers below.
Enjoyed this brief? Get the next one in your inbox.
More in Hardware & Compute
Dharma-AI reports higher GPU use with allocator
A constraint-aware scheduler beat FIFO in benchmarks on identical clusters.
Nvidia invests $1.5B in SoftBank data center developer
The investment will guarantee Nvidia chips power an OpenAI data center project.
Kog says GPUs can handle agentic workflows
French startup Kog argues deeper inference optimization can extract more from GPUs.