OpenAI's new chip allegedly 2x better perf/watt than Nvidia's Rubin
nickbaumann_ · x · 2026-08-26
SemiAnalysis reports that OpenAI's newest in-house chip achieves up to 2x better performance per watt than Nvidia's July Rubin performance. Notably, OpenAI's chip achieved this without even enabling speculative decoding, yet still beat the Rubin NVL72 configuration with it enabled.
More from Infra
- Alibaba's RecGPT-Mobile-V2: On-Device Behavior Prediction with RL — _reachsumit · 2026-08-26
- AMD MI350X Runs Qwen3.6-35B: Open Source Kernel Achieves 78.5k tok/s on 8 GPUs — SmilingGen · 2026-08-26
- MetricFire releases MCP server to query monitoring data with AI tools — PKMNPinBoard · 2026-08-26
- How continuous batching keeps GPUs busy: LLM inference runs steps, not requests — arpit_bhayani · 2026-08-26
- 30 Days of Inference: Deep dive into Blackwell and CUDA — blelbach · 2026-08-26
- Frequent Claude outages in August spark user concerns over model degradation — 新智元 · 2026-08-26