Wednesday, August 26, 2026HotTea verified storyVerified 10:44 AM PDT
← Back to the Wednesday, August 26, 2026 edition

OpenAI posts Jalapeño benchmarks before production ramps up

OpenAI published Jalapeño's first results on August 25. Across three public models, it reported 1.5 to 1.9 times more AI work per watt. It also reported 1.7 to 3.6 times lower end-to-end latency. OpenAI plans to deploy the chip internally by year end.

Verified 10:44 AM PDT · 2 original sources

The evidence

What the reporting establishes

What happened

OpenAI published Jalapeño's first results on August 25. Across three public models, it reported 1.5 to 1.9 times more AI work per watt. It also reported 1.7 to 3.6 times lower end-to-end latency. OpenAI plans to deploy the chip internally by year end.

Pressure point

These are company benchmarks. SemiAnalysis watched runs in OpenAI's lab and described the chips as engineering samples. It said production is scheduled to ramp up in 2027 and warned that the comparisons need caveats.

What to watch

OpenAI plans to start using Jalapeño internally by year end. Production volumes in 2027 and independent tests will show whether the lab results hold outside OpenAI's own testing.

Audit the story

Original sources

Company claims remain company claims. Follow the reporting and judge the evidence directly.

  1. OpenAIJalapeño first results
  2. SemiAnalysisOpenAI Jalapeño benchmark analysis

Continue the morning

Five stories. One sourced briefing.

Read the full editionListen to the daily audio →