OpenAI

GPT-5.1 Thinking

Multimodal

by OpenAI

+
+
+
+
About

GPT-5.1 Thinking is the extended reasoning variant of the GPT-5.1 family, applying deeper chain-of-thought processing to complex problems while being faster and delivering more understandable intermediate steps than its predecessor. It reflects OpenAI's ongoing refinement of the thinking model category — making extended reasoning both more accessible and more transparent.

+
+
+
+
Timeline
ReleasedNov 12, 2025
+
+
+
+
Specifications
Capabilities
Multimodal
+
+
+
+
License & Family
License
Proprietary
Performance Overview
Performance metrics and category breakdown

Overall Performance

2 benchmarks
Average Score
50.0%
Best Score
76.3%
High Performers (80%+)
0

Top Categories

Coding
76.3%
Reasoning
23.7%
+
+
+
+
All Benchmark Results for GPT-5.1 Thinking
Complete list of benchmark scores with detailed information
SWE Bench Verified
Coding
76.30
76.3%
Unverified
Humanity's Last Exam
Reasoning
23.68
23.7%
Unverified
+
+
+
+
Resources