Model comparison
GPT-5 Mini vs Mistral Large
GPT-5 Mini is the stronger model overall, scoring 41.7 to 31.6 on the Noometry Index.
Last verified . 38 shared benchmarks.
Mistral AISummary
- They share 38 benchmarks with published results for both. GPT-5 Mini scores higher in 9 categories and Mistral Large in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5 Mini leads 46.3 to 17.1.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 86.7% for GPT-5 Mini and 8.5% for Mistral Large.
- GPT-5 Mini is cheaper at $0.25 / $2 per million input/output tokens, against $2 / $6 for Mistral Large.
- GPT-5 Mini accepts more context: 400K tokens versus 131K.
- Mistral Large has downloadable open weights; the other is API-only.
Side by side
| GPT-5 Mini | Mistral Large | |
|---|---|---|
| Provider | OpenAI | Mistral AI |
| Noometry Index | 41.7 | 31.6 |
| Released | 2025-08-07 | 2024-02-26 |
| Weights | Proprietary | Open |
| Context window | 400K | 131K |
| Max output | 128K | 16K |
| Input $ / M tokens | $0.25 | $2 |
| Output $ / M tokens | $2 | $6 |
| Results tracked | 60 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5 Mini leads
GPT-5 Mini: 40.0 (#134), Mistral Large: 33.8 (#240)
| Benchmark | GPT-5 Mini | Mistral Large |
|---|---|---|
| SciCode | 39.2% | 36.2% |
| LMArena Coding | 1406 | 1277 |
| ALE-Bench | 799.77 | 264.7 |
| SWE-bench Verified | 64.7% | — |
| SWE-bench Verified (bash only) | 59.8% | — |
| SWE-bench Multilingual | 39.7% | — |
| WeirdML | 52.7% | — |
| BigCodeBench Instruct | — | 30% |
| LiveBench Coding | — | 47.1% |
| BigCodeBench Complete | — | 38.3% |
| AlgoTune | 1.38 | — |
| HumanEval+ | — | 62.2% |
| MBPP+ | — | 59.5% |
Agentic & Tool Use GPT-5 Mini leads
GPT-5 Mini: 31.0 (#64), Mistral Large: 28.4 (#83)
| Benchmark | GPT-5 Mini | Mistral Large |
|---|---|---|
| Berkeley Function Calling Leaderboard | 55.5% | 38.4% |
| Terminal-Bench | 34.8% | — |
| Vending-Bench 2 | -31.18 | — |
Reasoning GPT-5 Mini leads
GPT-5 Mini: 24.1 (#153), Mistral Large: 15.4 (#305)
| Benchmark | GPT-5 Mini | Mistral Large |
|---|---|---|
| CritPt | 0% | 0% |
| LMArena Hard Prompts | 1380 | 1257 |
| DTBench | 80.5% | 65.1% |
| LMCA | 34.2% | 16.7% |
| Epoch Capabilities Index | 145.51 | 128.52 |
| ForecastBench | 61 | 57.1 |
| ARC-AGI-2 | 4.4% | — |
| SimpleBench | — | 22.5% |
| Kagi LLM Benchmark | 70.3% | — |
| ARC-AGI-1 | 54.3% | — |
| Chess Puzzles | 30% | — |
| EnigmaEval | 8.2% | — |
| LiveBench Reasoning | — | 43.5% |
| Mystery Game Puzzles | 10% | — |
| LiveBench Data Analysis | — | 50.1% |
| LiveBench | — | 48.4% |
Math GPT-5 Mini leads
GPT-5 Mini: 46.3 (#59), Mistral Large: 17.1 (#285)
| Benchmark | GPT-5 Mini | Mistral Large |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 86.7% | 8.5% |
| Omni-MATH | 72.2% | 28.1% |
| LMArena Math | 1377 | 1262 |
| MATH Level 5 | 97.8% | 50.3% |
| FrontierMath (Feb 2025 set) | 27.2% | 0.3% |
| FrontierMath (Tiers 1-3) | 46.7% | — |
| FrontierMath Tier 4 | 12.2% | — |
| ProofBench | 9% | — |
| LiveBench Math | — | 42.5% |
| FrontierMath Tier 4 (v1) | 6.3% | — |
Knowledge GPT-5 Mini leads
GPT-5 Mini: 45.5 (#80), Mistral Large: 30.0 (#224)
| Benchmark | GPT-5 Mini | Mistral Large |
|---|---|---|
| GPQA Diamond | 75% | 51.3% |
| MMLU-Pro | 83.5% | 59.9% |
| Confabulations | 13.3% | 21.4% |
| Vectara Hallucination Rate | 12.9% | 4.5% |
| GPQA (HELM) | 75.6% | 43.5% |
| LMArena Expert | 1377 | 1232 |
| Humanity's Last Exam | 19.4% | — |
| SimpleQA Verified | 21.6% | — |
| MMLU | — | 80% |
Multimodal Not comparable
GPT-5 Mini: 35.0 (#80), Mistral Large: —
| Benchmark | GPT-5 Mini | Mistral Large |
|---|---|---|
| LMArena Vision | 1202 | — |
| VPCT | 40.2% | — |
Multilingual GPT-5 Mini leads
GPT-5 Mini: 49.1 (#131), Mistral Large: 40.1 (#214)
| Benchmark | GPT-5 Mini | Mistral Large |
|---|---|---|
| LMArena Non-English | 1363 | 1237 |
| LMArena Chinese | 1387 | 1240 |
| LMArena French | 1385 | 1324 |
| LMArena German | 1366 | 1254 |
| LMArena Japanese | 1342 | 1187 |
| LMArena Korean | 1306 | 1202 |
| LMArena Russian | 1362 | 1257 |
| LMArena Spanish | 1355 | 1268 |
Instruction Following GPT-5 Mini leads
GPT-5 Mini: 76.4 (#43), Mistral Large: 68.1 (#186)
| Benchmark | GPT-5 Mini | Mistral Large |
|---|---|---|
| IFEval | 92.7% | 87.7% |
| LMArena Instruction Following | 1357 | 1249 |
| LiveBench Instruction Following | — | 67.9% |
Long Context GPT-5 Mini leads
GPT-5 Mini: 41.9 (#127), Mistral Large: 38.3 (#194)
| Benchmark | GPT-5 Mini | Mistral Large |
|---|---|---|
| LMArena Longer Query | 1355 | 1262 |
| Fiction.LiveBench | 69.4% | — |
Writing & Preference GPT-5 Mini leads
GPT-5 Mini: 55.4 (#143), Mistral Large: 40.9 (#237)
| Benchmark | GPT-5 Mini | Mistral Large |
|---|---|---|
| LMArena Text | 1373 | 1266 |
| LMArena Creative Writing | 1325 | 1243 |
| Short-Story Creative Writing | 83.1% | 69% |
| EQ-Bench Creative Writing | 1313 | 985 |
| WildBench | 85.5% | 80.1% |
| LMArena Multi-Turn | 1363 | 1260 |
| LiveBench Language | — | 39.4% |
Frequently asked questions
Is GPT-5 Mini better than Mistral Large?
GPT-5 Mini is the stronger model overall, scoring 41.7 to 31.6 on the Noometry Index.
Which is cheaper, GPT-5 Mini or Mistral Large?
GPT-5 Mini is cheaper. It lists at $0.25 per million input tokens and $2 per million output tokens; Mistral Large lists at $2 and $6.
Is GPT-5 Mini or Mistral Large better for coding?
GPT-5 Mini scores higher on coding benchmarks: 40.0 versus 33.8 in the Noometry coding category.
Which has the bigger context window?
GPT-5 Mini does, with 400K tokens against 131K.
How many benchmarks do GPT-5 Mini and Mistral Large share?
38 benchmarks have published results for both models. GPT-5 Mini has 60 scored results on Noometry and Mistral Large has 51.