GPT-5.2 shipped as a later generation; this profile remains specific to the original gpt-5 model

The OnAirToday AI Progress Tracker recorded a verified change to OpenAI's GPT-5 (version gpt-5): GPT-5.2 shipped as a later generation; this profile remains specific to the original gpt-5 model.

GPT-5 is OpenAI’s August 2025 reasoning model for coding, mathematics, image understanding, and tool-assisted workflows, distinct from later GPT-5 generations. Published API specifications list a 400,000-token context window, text and image input, text output, and standard pricing of $1.25 per million input tokens and $10 per million output tokens.

What changed in this review:

• GPQA Diamond now reflects OpenAI’s published 85.7% result.
• SWE-bench Verified now reflects OpenAI’s published 74.9% result.
• Added OpenAI’s published AIME 2025 result of 94.6% without tools and MMMU result of 84.2%.
• Aider Polyglot remains 88%, with OpenAI’s release post as the result source.
• Terminal-Bench, LMArena, and other previously listed figures are not retained as confirmed results for this profile.
• Flaws and limitations updated

Teams building on GPT-5 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.

References:

• https://platform.openai.com/docs/models/gpt-5
• https://openai.com/index/introducing-gpt-5/
• https://openai.com/index/gpt-5-system-card/
• https://openai.com/index/introducing-gpt-5-2/