OpenAI’s Jalapeño chip earns TIME’s Best Inventions 2026 honor
OpenAI’s custom AI chip Jalapeño made TIME’s Best Inventions of 2026 list on October 15, marking its first major public recognition since launch. The chip, developed with Broadcom and manufactured by Celestica, is OpenAI’s first in-house Intelligence Processor—a custom ASIC optimized for inference tasks. It delivers 13.4 petaflops of 4-bit compute, 232 GB of memory, and 15.4 TB/s bandwidth, with…
Key points
- OpenAI’s Jalapeño chip earns TIME’s Best Inventions 2026 honor after launch in June 2026
- Delivers 13.4 petaflops, 232 GB memory, and 1.5–1.9x better AI efficiency per watt than Nvidia
- Small-volume deployment begins late 2026, targeting 10 GW capacity by 2029 with Broadcom and Celestica
The project ties to OpenAI’s broader hardware push, announced in October 2025, with Sam Altman, Richard Ho (OpenAI hardware VP), and Broadcom CEO Hock Tan leading the effort. OpenAI aims to hit 10 gigawatts of capacity by 2029, with Jalapeño’s small-volume deployment starting late 2026 and scaling in 2027. The chip targets 50% improvements in cost-per-token and per-kilowatt efficiency, though OpenAI has not disclosed exact cost savings. While Jalapeño is not a GPU replacement, it reflects a growing trend among tech giants—including Google, Amazon, and Microsoft—to build custom silicon for AI workloads, reducing reliance on Nvidia’s dominance in the space.
The story so far
4 episodes →- OpenAI’s Jalapeño chip earns TIME’s Best Inventions 2026 honorthis story
OpenAI’s Jalapeño chip lands on TIME’s Best Inventions of 2026 list
cryptobriefing.com · 16 October 2026
Loading the full article…
This text was published by cryptobriefing.com and written by Diego Almada Lopez. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Hardware & Compute
All →- Nvidia’s H100 gpu worth about 30% more than a year ago · 1 src
- Nvidia, AMD, Micron drop premarket as AI financing fears rise · 5 src
- Nvidia launches RTX Spark platform with Microsoft Surface Laptop Ultra featuring 20-core CPU and 6144 CUDA cores · 16 src
- NVIDIA launches RTX Spark PC for creators and AI agents · 2 src
- Thursday, October 15: Anthropic to Build a Data Center in Osaka Worth Up to 1 Trillion Yen—Is the Massive AI Investment Really Worth the Returns? · 1 src
Comments
via GitHub Discussions