Rankings / OpenAI
GPT 5.1 Codex Mini
released 2025-11-13
BenchAtlas Index
as of 2026-07-14
high effort
rank #152
9 families · 5 categories · high
Benchmark evidence
11 results
agentic coding
| τ²-Bench | high effort | 62.9 % | independent | Artificial Analysis |
coding
| LiveCodeBench v6 implementation=artificial-analysis | high effort | 83.6 % | independent | Artificial Analysis |
| SciCode | high effort | 42.6 % | independent | Artificial Analysis |
external indices
| Intelligence Index v4.1 | high effort | 30.6 points | independent | Artificial Analysis |
| AA Math Index | high effort | 91.7 points | independent | Artificial Analysis |
knowledge science
| GPQA Diamond implementation=artificial-analysis | high effort | 81.3 % | independent | Artificial Analysis |
| Humanity's Last Exam implementation=artificial-analysis | high effort | 16.9 % | independent | Artificial Analysis |
| MMLU-Pro implementation=artificial-analysis | high effort | 82.0 % | independent | Artificial Analysis |
long context instruction
| AA-LCR | high effort | 62.7 % | independent | Artificial Analysis |
| IFBench | high effort | 67.9 % | independent | Artificial Analysis |
reasoning math
| AIME year=2025 · implementation=artificial-analysis | high effort | 91.7 % | independent | Artificial Analysis |
Agent + model results
systems, not bare-model scores
| agent + model hookele + GPT 5.1 Codex Mini | Terminal-Bench 2.0 | 61.6 % | unverified | Terminal-Bench Leaderboard |
| agent + model Crux + GPT 5.1 Codex Mini | Terminal-Bench 2.0 | 43.1 % | community | Terminal-Bench Leaderboard |
| agent + model Artificial Analysis harness + GPT 5.1 Codex Mini | Terminal-Bench Hard | 33.3 % | independent | Artificial Analysis |
These scores measure the whole agent system (scaffold, tools, budgets) — they are never merged into the bare model’s numbers.