OpenAI's first custom chip "Jalapeño" reportedly beats Nvidia's Blackwell and Rubin in inference benchmarks
The Decoderen

OpenAI showed off "Jalapeño," its first in-house inference chip, with benchmarks at the Hot Chips conference. According to SemiAnalysis tests, the chip beats Nvidia's Blackwell and even Rubin in throughput and energy efficiency. "Usually first generation chips aren't competitive, but OpenAI is beating Nvidia Blackwell and even Rubin," SemiAnalysis CEO Dylan Patel said. The article OpenAI's first custom chip "Jalapeño" reportedly beats Nvidia's Blackwell and Rubin in inference benchmarks appeared first on The Decoder .
This is a short summary published by AI Global Wire. The full article is owned and hosted by The Decoder — open it there to read it in full.
Read the full story at The Decoder- OpenAI
- Verktyg
Related AI news
- Sources: Chris Malone, who joined OpenAI as head of data centers in March 2025, left the company last week amid a broader exodus (Anissa Gardizy/Wall Street Journal)Techmeme · August 25, 2026
- AI inference gets a new tier as context windows growSiliconANGLE · August 25, 2026
- Agentic observability with Amazon OpenSearch Service MCP AppsAWS Machine Learning · August 25, 2026
- Russia used ChatGPT to run a covert influence campaign pushing pro-Kremlin narratives across the WestThe Decoder · August 25, 2026
- Democratizing the AI data center: How Cisco and Nvidia are bringing rack-scale power to the enterpriseSiliconANGLE · August 25, 2026
- Prisjakt återupptar börsplanerna: ”Vi är redo”DI Digital · August 25, 2026