OpenAI's Jalapeño Chip Beats Nvidia Blackwell in Early Benchmarks
itsclivetime · x · 2026-08-26
SemiAnalysis reports that OpenAI's in-house inference chip, Jalapeño, has outperformed Nvidia's VR200 in A0 silicon testing with faster program execution. Developed with Broadcom, the chip is designed for LLM inference and utilizes a non-SpecDec implementation of DeepSeek. The report highlights an aggressive 16-month design cycle and notes that Jalapeño achieves high performance across various scenarios through hardware-software co-design, surpassing current chips from Nvidia, AMD, and Google in multiple tests.
More from Infra
- engy.ai processes 50B+ tokens in a day as open-model inference demand surges — const_reborn · 2026-10-16
- New inference engine 'e' hits 10,500 tok/s on a single DGX Station — brandon_galang · 2026-10-16
- SGLang ships Day-0 support for FLUX 3 Image Klein: 10 reference images, BF16/FP8/NVFP4 variants — ying11231 · 2026-10-16
- NVIDIA's Nemotron 3.5 Super VL hits CoreWeave: 12B-active multimodal MoE at $0.25/M input tokens — wandb · 2026-10-16
- Chip equipment beats bull case: 2026 WFE seen at $153B, $278B by 2030 — BenBajarin · 2026-10-16
- Steam now launches in 5.23s, beating Epic's 5.47s — and vLLM gets roasted for slow startup — AlpinDale · 2026-10-16