GPT-5.1: added launch-reported SWE-bench Verified result of 76.3%; this is historical evidence, not a newly measured improvement
Quantifying the autonomous coding capabilities of the latest model iteration.
Added launch-reported SWE-bench Verified result of 76.3%; this is historical evidence, not a newly measured improvement — verified by the OnAirToday AI Progress Tracker against OpenAI's published documentation.
OnAirToday
Newsroom
The OnAirToday AI Progress Tracker recorded a verified change to OpenAI's GPT-5.1 (version gpt-5.1): Added launch-reported SWE-bench Verified result of 76.3%; this is historical evidence, not a newly measured improvement.
GPT-5.1 is an OpenAI model for text generation, coding and tool-assisted workflows, with image input and configurable reasoning effort. This profile reflects published launch documentation rather than a live-verified September 2026 status; current availability, pricing changes and successor releases could not be independently checked here.
Teams building on GPT-5.1 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-5-1.
References:
- https://platform.openai.com/docs/models/gpt-5.1
- https://openai.com/index/gpt-5-1-for-developers/
- https://openai.com/index/gpt-5-1/
Generated insight by OnAirToday AI
The documented 76.3% SWE-bench Verified score establishes a critical performance baseline for autonomous software engineering capabilities. This milestone confirms that LLMs are increasingly capable of resolving complex, real-world coding tasks with professional-grade accuracy.
Comments
Keep comments relevant and respectful. No links, advertising, self-promotion, or spam.