Model comparison
GPT-5 Mini vs Mistral Small
GPT-5 Mini is the stronger model overall, scoring 41.8 to 33.4 on the Noometry Index. Mistral Small costs 2.6× less per token, which makes it the better buy when GPT-5 Mini's lead doesn't matter for your workload.
Last verified . 29 shared benchmarks.
Mistral AISummary
- They share 29 benchmarks with published results for both. GPT-5 Mini scores higher in 10 categories and Mistral Small in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5 Mini leads 46.7 to 16.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 86.7% for GPT-5 Mini and 5.8% for Mistral Small.
- Mistral Small is cheaper at $0.15 / $0.60 per million input/output tokens, against $0.25 / $2 for GPT-5 Mini.
- GPT-5 Mini accepts more context: 400K tokens versus 262K.
- Mistral Small has downloadable open weights; the other is API-only.
Side by side
| GPT-5 Mini | Mistral Small | |
|---|---|---|
| Provider | OpenAI | Mistral AI |
| Noometry Index | 41.8 | 33.4 |
| Released | 2025-08-07 | 2024-02-26 |
| Weights | Proprietary | Open |
| Context window | 400K | 262K |
| Max output | 128K | 256K |
| Input $ / M tokens | $0.25 | $0.15 |
| Output $ / M tokens | $2 | $0.60 |
| Results tracked | 60 | 39 |
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 Small: 34.0 (#247)
| Benchmark | GPT-5 Mini | Mistral Small |
|---|---|---|
| SciCode | 39.2% | 26.5% |
| LMArena Coding | 1406 | 1362 |
| ALE-Bench | 799.77 | 497.62 |
| SWE-bench Verified | 64.7% | — |
| SWE-bench Verified (bash only) | 59.8% | — |
| SWE-bench Multilingual | 39.7% | — |
| WeirdML | 52.7% | — |
| BigCodeBench Instruct | — | 36.1% |
| LiveBench Coding | — | 36.2% |
| BigCodeBench Complete | — | 46.6% |
| AlgoTune | 1.38 | — |
Agentic & Tool Use GPT-5 Mini leads
GPT-5 Mini: 31.1 (#70), Mistral Small: 28.1 (#93)
| Benchmark | GPT-5 Mini | Mistral Small |
|---|---|---|
| Berkeley Function Calling Leaderboard | 55.5% | 37.1% |
| Terminal-Bench | 34.8% | — |
| Vending-Bench 2 | -31.18 | — |
Reasoning GPT-5 Mini leads
GPT-5 Mini: 23.9 (#168), Mistral Small: 19.8 (#250)
| Benchmark | GPT-5 Mini | Mistral Small |
|---|---|---|
| Kagi LLM Benchmark | 70.3% | 37.8% |
| CritPt | 0% | 0% |
| LMArena Hard Prompts | 1380 | 1335 |
| DTBench | 80.5% | 70.9% |
| LMCA | 34.2% | 20.6% |
| ARC-AGI-2 | 4.4% | — |
| ARC-AGI-1 | 54.3% | — |
| Chess Puzzles | 30% | — |
| EnigmaEval | 8.2% | — |
| LiveBench Reasoning | — | 44.8% |
| Mystery Game Puzzles | 10% | — |
| LiveBench Data Analysis | — | 53.7% |
| Epoch Capabilities Index | 145.52 | — |
| ForecastBench | 61 | — |
| LiveBench | — | 44% |
Math GPT-5 Mini leads
GPT-5 Mini: 46.7 (#69), Mistral Small: 16.4 (#293)
| Benchmark | GPT-5 Mini | Mistral Small |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 86.7% | 5.8% |
| LMArena Math | 1378 | 1341 |
| MATH Level 5 | 97.8% | 46.8% |
| FrontierMath (Tiers 1-3) | 46.7% | — |
| FrontierMath Tier 4 | 12.2% | — |
| ProofBench | 9% | — |
| Omni-MATH | 72.2% | — |
| LiveBench Math | — | 39.9% |
| FrontierMath (Feb 2025 set) | 27.2% | — |
| FrontierMath Tier 4 (v1) | 6.3% | — |
Knowledge GPT-5 Mini leads
GPT-5 Mini: 45.6 (#86), Mistral Small: 31.0 (#222)
| Benchmark | GPT-5 Mini | Mistral Small |
|---|---|---|
| GPQA Diamond | 75% | 47.5% |
| Vectara Hallucination Rate | 12.9% | 5.1% |
| LMArena Expert | 1379 | 1291 |
| Humanity's Last Exam | 19.4% | — |
| SimpleQA Verified | 21.6% | — |
| MMLU-Pro | 83.5% | — |
| Confabulations | 13.3% | — |
| GPQA (HELM) | 75.6% | — |
| MMLU | — | 68.7% |
Multimodal GPT-5 Mini leads
GPT-5 Mini: 35.6 (#85), Mistral Small: 33.5 (#96)
| Benchmark | GPT-5 Mini | Mistral Small |
|---|---|---|
| LMArena Vision | 1202 | 1142 |
| VPCT | 40.2% | — |
Multilingual GPT-5 Mini leads
GPT-5 Mini: 48.9 (#137), Mistral Small: 45.5 (#169)
| Benchmark | GPT-5 Mini | Mistral Small |
|---|---|---|
| LMArena Non-English | 1363 | 1315 |
| LMArena Chinese | 1385 | 1340 |
| LMArena French | 1386 | 1337 |
| LMArena German | 1366 | 1340 |
| LMArena Japanese | 1341 | 1275 |
| LMArena Korean | 1308 | 1259 |
| LMArena Russian | 1362 | 1324 |
| LMArena Spanish | 1355 | 1346 |
Instruction Following GPT-5 Mini leads
GPT-5 Mini: 76.2 (#46), Mistral Small: 66.4 (#209)
| Benchmark | GPT-5 Mini | Mistral Small |
|---|---|---|
| LMArena Instruction Following | 1357 | 1310 |
| LiveBench Instruction Following | — | 63.7% |
| IFEval | 92.7% | — |
Long Context GPT-5 Mini leads
GPT-5 Mini: 41.9 (#132), Mistral Small: 40.4 (#156)
| Benchmark | GPT-5 Mini | Mistral Small |
|---|---|---|
| LMArena Longer Query | 1355 | 1327 |
| Fiction.LiveBench | 69.4% | — |
Writing & Preference GPT-5 Mini leads
GPT-5 Mini: 55.2 (#148), Mistral Small: 52.5 (#171)
| Benchmark | GPT-5 Mini | Mistral Small |
|---|---|---|
| LMArena Text | 1373 | 1338 |
| LMArena Creative Writing | 1325 | 1305 |
| LMArena Multi-Turn | 1363 | 1344 |
| Short-Story Creative Writing | 83.1% | — |
| EQ-Bench Creative Writing | 1313 | — |
| WildBench | 85.5% | — |
| LiveBench Language | — | 30.5% |
Frequently asked questions
Is GPT-5 Mini better than Mistral Small?
GPT-5 Mini is the stronger model overall, scoring 41.8 to 33.4 on the Noometry Index. Mistral Small costs 2.6× less per token, which makes it the better buy when GPT-5 Mini's lead doesn't matter for your workload.
Which is cheaper, GPT-5 Mini or Mistral Small?
Mistral Small is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; GPT-5 Mini lists at $0.25 and $2.
Is GPT-5 Mini or Mistral Small better for coding?
GPT-5 Mini scores higher on coding benchmarks: 40.1 versus 34.0 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 Small share?
29 benchmarks have published results for both models. GPT-5 Mini has 60 scored results on Noometry and Mistral Small has 39.