GPT-5-Mini: gPQA Diamond and SWE-bench Verified entries now use OpenAI’s published results, rather than unattributed evaluation configurations

The OnAirToday AI Progress Tracker recorded a verified change to OpenAI's GPT-5-Mini (version gpt-5-mini): GPQA Diamond and SWE-bench Verified entries now use OpenAI’s published results, rather than unattributed evaluation configurations.

GPT-5 mini is OpenAI’s smaller GPT-5 reasoning model, designed for cost-efficient work on well-defined tasks. It accepts text and image inputs and produces text, with standard API pricing of $0.25 per million input tokens and $2 per million output tokens.

What changed in this review:

• Added OpenAI’s published AIME 2025 result.
• Documented reasoning controls, output limits, modality restrictions and deployment risks.
• 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://openai.com/index/gpt-5-system-card/
• https://platform.openai.com/docs/models/gpt-4o-mini