OpenAI Reports Jalapeño Chip Delivers Up to 1.9× More AI Work per Watt
Sign in✓ Signed in
OpenAI has published the first measured results for Jalapeño, its custom inference chip, reporting 1.5–1.9× more AI work per watt and 1.7–3.6× lower end-to-end latency than comparison systems. The results were measured on InferenceX, a public benchmark from SemiAnalysis that evaluates the full process of serving an AI request. OpenAI’s benchmark report covers GPT‑OSS 120B, DeepSeek R1 670B and Kimi K2.5 1T.
Continue reading
The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.
Already purchased? Sign in✓ Signed in — this article isn’t included in your current plan.
Unlocking the full article...