BREAKING
OpenAI, Broadcom Unveil 'Jalapeño' Chip
Built for LLM Inference, Not Training
0
mo
to tape-out
0
%
cost savings vs GPU
Full-Stack Strategy Roles
1
OpenAI design
↓
2
Broadcom silicon
↓
3
Celestica systems
Strengths vs Open Questions
Promised Gains
●
Better perf per watt
●
Lower inference cost
●
Less NVIDIA dependence
Caveats
●
No independent tests
●
Still lab stage
●
Specs undisclosed
Deployment Begins End of 2026
AI NEWS BLITZ
OpenAI and Broadcom have unveiled Jalapeño, OpenAI's first in-house AI chip.