SemiAnalysis Benchmarks Put OpenAI's Jalapeño Chip Ahead of Nvidia Blackwell on Efficiency
Summary
SemiAnalysis InferenceX benchmarks show OpenAI's Jalapeño inference chip beating Nvidia's Blackwell on performance per watt across models including GPT-OSS and DeepSeek R1. Ho, OpenAI's hardware vice president, says each of the chip's 64 cores addresses its own HBM bank, and the design reached tapeout in nine months.
Key Points
- Ho says about 100 people cover chip design, verification, packaging, thermals, rack design, supply chain, software, and firmware for Jalapeño.
- Ho says OpenAI chose HBM4 over SRAM because its large models need capacity that process nodes cannot supply.
- Ho says OpenAI has started work on second, third, and fourth generations of the chip.