Compare / Mistral Medium 3.1 vs GPT-5 Mini
Mistral Medium 3.1 vs GPT-5 Mini
21 results from 3 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | Mistral Medium 3.1 | GPT-5 Mini |
|---|---|---|---|
| Arena (formerly LMArena)reported | OverallArena rating | 1,429 | 1,394 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,407 | 1,381 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,378 | 1,324 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,410 | 1,400 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,395 | 1,373 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,408 | 1,390 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,386 | 1,352 |
| Artificial Analysisreported | Artificial Analysis Coding Indexindex score | 20.5 | 15.6 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 9.20 | 20.6 |
| Artificial Analysisreported | GPQA Diamond% of questions | 58.8% | 82.8% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 4.7% | 21.5% |
| Artificial Analysisreported | IFBench% of instructions | 39.8% | 75.4% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 42.7% | 72.3% |
| Artificial Analysisreported | SciCode% of problems | 32.4% | 39.0% |
| Artificial Analysisreported | Terminal-Bench 2.1% of tasks | 13.9% | 3.7% |
| Artificial Analysisreported | Terminal-Bench 4.0% of tasks | 0.0% | 0.0% |
| Artificial Analysisreported | Terminal-Bench Hard% of tasks | 10.6% | 33.3% |
| Artificial Analysisreported | Τ-bench banking% of tasks | 8.2% | 15.5% |
| Artificial Analysisreported | Τ²-bench telecom% of tasks | 40.6% | 71.1% |
| Vectara Hallucination Leaderboardreported | Answer rate% of documents | 99.7% | 99.9% |
| Vectara Hallucination Leaderboardreported | Hallucination rate% of summaries | 22.7% | 12.9% |
Bold green marks the better value on that metric. Where a source reports ranges that overlap, the difference may not be meaningful; see the benchmark page for ranges.