GPT-5-Mini: gPQA Diamond changes from 75% to OpenAI’s published 82.3%; this is a reporting correction, not evidence of a model upgrade

The OnAirToday AI Progress Tracker recorded a verified change to OpenAI's GPT-5-Mini (version gpt-5-mini): GPQA Diamond changes from 75% to OpenAI’s published 82.3%; this is a reporting correction, not evidence of a model upgrade.

GPT-5 mini is OpenAI’s smaller GPT-5 reasoning model for well-defined tasks, accepting text and image inputs and producing text outputs. It supports a 400,000-token context window, up to 128,000 output tokens, and standard API pricing of $0.25 per million input tokens and $2.00 per million output tokens.

What changed in this review:

• SWE-bench Verified changes from 64.7% to OpenAI’s published 71.0%; evaluation configurations may differ.
• Adds OpenAI’s published AIME 2025 result of 91.1% and MMMU result of 81.6%.
• Flaws and limitations updated
• Risks and threats updated

Teams building on GPT-5-Mini should confirm limits, pricing and benchmark coverage against the vendor's own documentation before relying on these figures.

The full profile — capabilities, benchmark citations, known flaws and risk guidance — is on the tracker at /ai-models/gpt-5-mini.

References:

• https://platform.openai.com/docs/models/gpt-5-mini
• https://openai.com/index/introducing-gpt-5-for-developers/
• https://platform.openai.com/docs/models/gpt-4o-mini
• https://platform.openai.com/docs/guides/reasoning