Model comparison
GPT-5 Mini vs Mistral Medium
GPT-5 Mini is the stronger model overall, scoring 41.8 to 36.3 on the Noometry Index.
Last verified . 34 shared benchmarks.
Mistral AISummary
- They share 34 benchmarks with published results for both. GPT-5 Mini scores higher in 6 categories and Mistral Medium in 4 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where GPT-5 Mini leads 45.6 to 25.0.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 86.7% for GPT-5 Mini and 32.2% for Mistral Medium.
- GPT-5 Mini is cheaper at $0.25 / $2 per million input/output tokens, against $1.50 / $7.50 for Mistral Medium.
- GPT-5 Mini accepts more context: 400K tokens versus 262K.
- Mistral Medium has downloadable open weights; the other is API-only.
Side by side
| GPT-5 Mini | Mistral Medium | |
|---|---|---|
| Provider | OpenAI | Mistral AI |
| Noometry Index | 41.8 | 36.3 |
| Released | 2025-08-07 | 2023-12-11 |
| Weights | Proprietary | Open |
| Context window | 400K | 262K |
| Max output | 128K | 262K |
| Input $ / M tokens | $0.25 | $1.50 |
| Output $ / M tokens | $2 | $7.50 |
| Results tracked | 60 | 36 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5 Mini leads
GPT-5 Mini: 40.1 (#146), Mistral Medium: 34.2 (#243)
| Benchmark | GPT-5 Mini | Mistral Medium |
|---|---|---|
| SciCode | 39.2% | 40.2% |
| WeirdML | 52.7% | 43.7% |
| LMArena Coding | 1406 | 1434 |
| ALE-Bench | 799.77 | 763.98 |
| SWE-bench Verified | 64.7% | — |
| FrontierCode | — | 8% |
| SWE-bench Verified (bash only) | 59.8% | — |
| SWE-bench Multilingual | 39.7% | — |
| AlgoTune | 1.38 | — |
Agentic & Tool Use GPT-5 Mini leads
GPT-5 Mini: 31.1 (#70), Mistral Medium: 28.3 (#90)
| Benchmark | GPT-5 Mini | Mistral Medium |
|---|---|---|
| Berkeley Function Calling Leaderboard | 55.5% | 37.7% |
| Terminal-Bench | 34.8% | — |
| Vending-Bench 2 | -31.18 | — |
Reasoning Too close to call
GPT-5 Mini: 23.9 (#168), Mistral Medium: 24.0 (#167)
| Benchmark | GPT-5 Mini | Mistral Medium |
|---|---|---|
| Kagi LLM Benchmark | 70.3% | 50% |
| CritPt | 0% | 0% |
| LMArena Hard Prompts | 1380 | 1426 |
| DTBench | 80.5% | 75.5% |
| LMCA | 34.2% | 26.1% |
| ARC-AGI-2 | 4.4% | — |
| ARC-AGI-1 | 54.3% | — |
| Chess Puzzles | 30% | — |
| EnigmaEval | 8.2% | — |
| Mystery Game Puzzles | 10% | — |
| Surface Evolver Bench | — | 26.9% |
| Epoch Capabilities Index | 145.52 | — |
| ForecastBench | 61 | — |
Math GPT-5 Mini leads
GPT-5 Mini: 46.7 (#69), Mistral Medium: 28.1 (#245)
| Benchmark | GPT-5 Mini | Mistral Medium |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 86.7% | 32.2% |
| ProofBench | 9% | 9% |
| LMArena Math | 1378 | 1408 |
| MATH Level 5 | 97.8% | 81.6% |
| FrontierMath (Feb 2025 set) | 27.2% | 0.3% |
| FrontierMath (Tiers 1-3) | 46.7% | — |
| FrontierMath Tier 4 | 12.2% | — |
| Omni-MATH | 72.2% | — |
| FrontierMath Tier 4 (v1) | 6.3% | — |
Knowledge GPT-5 Mini leads
GPT-5 Mini: 45.6 (#86), Mistral Medium: 25.0 (#265)
| Benchmark | GPT-5 Mini | Mistral Medium |
|---|---|---|
| GPQA Diamond | 75% | 59.5% |
| Humanity's Last Exam | 19.4% | 4.5% |
| Vectara Hallucination Rate | 12.9% | 22.7% |
| LMArena Expert | 1379 | 1408 |
| SimpleQA Verified | 21.6% | — |
| MMLU-Pro | 83.5% | — |
| Confabulations | 13.3% | — |
| GPQA (HELM) | 75.6% | — |
Multimodal Too close to call
GPT-5 Mini: 35.6 (#85), Mistral Medium: 35.3 (#88)
| Benchmark | GPT-5 Mini | Mistral Medium |
|---|---|---|
| LMArena Vision | 1202 | 1172 |
| VPCT | 40.2% | — |
Multilingual Mistral Medium leads
GPT-5 Mini: 48.9 (#137), Mistral Medium: 52.1 (#91)
| Benchmark | GPT-5 Mini | Mistral Medium |
|---|---|---|
| LMArena Non-English | 1363 | 1408 |
| LMArena Chinese | 1385 | 1447 |
| LMArena French | 1386 | 1459 |
| LMArena German | 1366 | 1432 |
| LMArena Japanese | 1341 | 1378 |
| LMArena Korean | 1308 | 1380 |
| LMArena Russian | 1362 | 1411 |
| LMArena Spanish | 1355 | 1433 |
Instruction Following GPT-5 Mini leads
GPT-5 Mini: 76.2 (#46), Mistral Medium: 73.7 (#116)
| Benchmark | GPT-5 Mini | Mistral Medium |
|---|---|---|
| LMArena Instruction Following | 1357 | 1398 |
| IFEval | 92.7% | — |
Long Context Mistral Medium leads
GPT-5 Mini: 41.9 (#132), Mistral Medium: 42.9 (#114)
| Benchmark | GPT-5 Mini | Mistral Medium |
|---|---|---|
| LMArena Longer Query | 1355 | 1406 |
| Fiction.LiveBench | 69.4% | — |
Writing & Preference Mistral Medium leads
GPT-5 Mini: 55.2 (#148), Mistral Medium: 60.0 (#103)
| Benchmark | GPT-5 Mini | Mistral Medium |
|---|---|---|
| LMArena Text | 1373 | 1424 |
| LMArena Creative Writing | 1325 | 1391 |
| Short-Story Creative Writing | 83.1% | 77.3% |
| LMArena Multi-Turn | 1363 | 1418 |
| EQ-Bench Creative Writing | 1313 | — |
| WildBench | 85.5% | — |
Frequently asked questions
Is GPT-5 Mini better than Mistral Medium?
GPT-5 Mini is the stronger model overall, scoring 41.8 to 36.3 on the Noometry Index.
Which is cheaper, GPT-5 Mini or Mistral Medium?
GPT-5 Mini is cheaper. It lists at $0.25 per million input tokens and $2 per million output tokens; Mistral Medium lists at $1.50 and $7.50.
Is GPT-5 Mini or Mistral Medium better for coding?
GPT-5 Mini scores higher on coding benchmarks: 40.1 versus 34.2 in the Noometry coding category.
Which has the bigger context window?
GPT-5 Mini does, with 400K tokens against 262K.
How many benchmarks do GPT-5 Mini and Mistral Medium share?
34 benchmarks have published results for both models. GPT-5 Mini has 60 scored results on Noometry and Mistral Medium has 36.