OpenAI's Gen 1 Chip Challenges Nvidia Vera Rubin in Inference Efficiency
haider1 · x · 2026-08-26
SemiAnalysis' InferenceX benchmarks reveal that OpenAI's first-generation chip, "Jalapeño," delivers impressive inference performance, challenging Nvidia's upcoming Vera Rubin.
Benchmark Highlights
- DeepSeek R1: Jalapeño achieves 1.7x better performance-per-watt and 3.6x lower latency than Rubin, significantly outperforming GB200.
- GPT-OSS: 1.9x better performance-per-watt and 1.7x lower latency.
- Kimi K2.5: 1.5x better performance-per-watt and 3.4x lower latency.
First-generation chips are rarely competitive, but OpenAI's entry beats Nvidia Blackwell and rivals the next-gen Rubin in key metrics.
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