Model comparison
GPT-5 vs Mistral Small
GPT-5 is the stronger model overall, scoring 50.9 to 33.4 on the Noometry Index. Mistral Small costs 13× less per token, which makes it the better buy when GPT-5's lead doesn't matter for your workload.
Last verified . 28 shared benchmarks.
Mistral AISummary
- They share 28 benchmarks with published results for both. GPT-5 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 leads 55.0 to 16.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 91.4% for GPT-5 and 5.8% for Mistral Small.
- Mistral Small is cheaper at $0.15 / $0.60 per million input/output tokens, against $1.25 / $10 for GPT-5.
- GPT-5 accepts more context: 400K tokens versus 262K.
- Mistral Small has downloadable open weights; the other is API-only.
Side by side
| GPT-5 | Mistral Small | |
|---|---|---|
| Provider | OpenAI | Mistral AI |
| Noometry Index | 50.9 | 33.4 |
| Released | 2025-08-07 | 2024-02-26 |
| Weights | Proprietary | Open |
| Context window | 400K | 262K |
| Max output | 128K | 256K |
| Input $ / M tokens | $1.25 | $0.15 |
| Output $ / M tokens | $10 | $0.60 |
| Results tracked | 69 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5 leads
GPT-5: 50.3 (#47), Mistral Small: 34.0 (#247)
| Benchmark | GPT-5 | Mistral Small |
|---|---|---|
| SciCode | 42.9% | 26.5% |
| LMArena Coding | 1436 | 1362 |
| ALE-Bench | 1,162 | 497.62 |
| SWE-bench Verified | 73.6% | — |
| SWE-bench Verified (bash only) | 65% | — |
| Aider Polyglot | 88% | — |
| LMArena WebDev | 1418 | — |
| GSO | 6.9% | — |
| WeirdML | 60.7% | — |
| BigCodeBench Instruct | — | 36.1% |
| LiveBench Coding | — | 36.2% |
| BigCodeBench Complete | — | 46.6% |
| AlgoTune | 1.67 | — |
Agentic & Tool Use GPT-5 leads
GPT-5: 33.1 (#56), Mistral Small: 28.1 (#93)
| Benchmark | GPT-5 | Mistral Small |
|---|---|---|
| Terminal-Bench | 49.6% | — |
| Berkeley Function Calling Leaderboard | — | 37.1% |
| GDPval | 34.8% | — |
| Remote Labor Index | 1.7% | — |
| DeepResearch Bench | 49.6% | — |
| BALROG | 32.8% | — |
| LMArena Search | 1133 | — |
| METR Time Horizons | 69.6% | — |
Reasoning GPT-5 leads
GPT-5: 38.3 (#64), Mistral Small: 19.8 (#250)
| Benchmark | GPT-5 | Mistral Small |
|---|---|---|
| Kagi LLM Benchmark | 72.7% | 37.8% |
| CritPt | 12.6% | 0% |
| LMArena Hard Prompts | 1416 | 1335 |
| DTBench | 90.7% | 70.9% |
| LMCA | 40% | 20.6% |
| ARC-AGI-2 | 9.9% | — |
| SimpleBench | 56.7% | — |
| ARC-AGI-1 | 65.7% | — |
| Chess Puzzles | 37% | — |
| EnigmaEval | 10.5% | — |
| EBR-Bench | 12.7% | — |
| LiveBench Reasoning | — | 44.8% |
| Mystery Game Puzzles | 23% | — |
| LiveBench Data Analysis | — | 53.7% |
| Epoch Capabilities Index | 150 | — |
| ForecastBench | 61.4 | — |
| LiveBench | — | 44% |
Math GPT-5 leads
GPT-5: 55.0 (#44), Mistral Small: 16.4 (#293)
| Benchmark | GPT-5 | Mistral Small |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 91.4% | 5.8% |
| LMArena Math | 1407 | 1341 |
| MATH Level 5 | 98.1% | 46.8% |
| FrontierMath (Tiers 1-3) | 55.4% | — |
| FrontierMath Tier 4 | 22% | — |
| ProofBench | 18% | — |
| Omni-MATH | 64.7% | — |
| LiveBench Math | — | 39.9% |
| FrontierMath (Feb 2025 set) | 32.4% | — |
| FrontierMath Tier 4 (v1) | 12.5% | — |
Knowledge GPT-5 leads
GPT-5: 56.6 (#43), Mistral Small: 31.0 (#222)
| Benchmark | GPT-5 | Mistral Small |
|---|---|---|
| GPQA Diamond | 86.2% | 47.5% |
| Vectara Hallucination Rate | 14.7% | 5.1% |
| LMArena Expert | 1419 | 1291 |
| Humanity's Last Exam | 25.3% | — |
| SimpleQA Verified | 50.1% | — |
| MMLU-Pro | 86.3% | — |
| Confabulations | 10.3% | — |
| GPQA (HELM) | 79.2% | — |
| MMLU | — | 68.7% |
Multimodal GPT-5 leads
GPT-5: 46.8 (#13), Mistral Small: 33.5 (#96)
| Benchmark | GPT-5 | Mistral Small |
|---|---|---|
| LMArena Vision | 1232 | 1142 |
| GeoBench | 81% | — |
| VPCT | 66% | — |
Multilingual GPT-5 leads
GPT-5: 51.4 (#110), Mistral Small: 45.5 (#169)
| Benchmark | GPT-5 | Mistral Small |
|---|---|---|
| LMArena Non-English | 1397 | 1315 |
| LMArena Chinese | 1422 | 1340 |
| LMArena French | 1410 | 1337 |
| LMArena German | 1416 | 1340 |
| LMArena Japanese | 1409 | 1275 |
| LMArena Korean | 1360 | 1259 |
| LMArena Russian | 1406 | 1324 |
| LMArena Spanish | 1399 | 1346 |
Instruction Following GPT-5 leads
GPT-5: 73.8 (#113), Mistral Small: 66.4 (#209)
| Benchmark | GPT-5 | Mistral Small |
|---|---|---|
| LMArena Instruction Following | 1388 | 1310 |
| LiveBench Instruction Following | — | 63.7% |
| IFEval | 87.5% | — |
Long Context GPT-5 leads
GPT-5: 69.5 (#2), Mistral Small: 40.4 (#156)
| Benchmark | GPT-5 | Mistral Small |
|---|---|---|
| LMArena Longer Query | 1399 | 1327 |
| Fiction.LiveBench | 97.2% | — |
Writing & Preference GPT-5 leads
GPT-5: 63.4 (#65), Mistral Small: 52.5 (#171)
| Benchmark | GPT-5 | Mistral Small |
|---|---|---|
| LMArena Text | 1406 | 1338 |
| LMArena Creative Writing | 1365 | 1305 |
| LMArena Multi-Turn | 1426 | 1344 |
| Short-Story Creative Writing | 86% | — |
| EQ-Bench Creative Writing | 1627 | — |
| WildBench | 85.7% | — |
| LiveBench Language | — | 30.5% |
Frequently asked questions
Is GPT-5 better than Mistral Small?
GPT-5 is the stronger model overall, scoring 50.9 to 33.4 on the Noometry Index. Mistral Small costs 13× less per token, which makes it the better buy when GPT-5's lead doesn't matter for your workload.
Which is cheaper, GPT-5 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 lists at $1.25 and $10.
Is GPT-5 or Mistral Small better for coding?
GPT-5 scores higher on coding benchmarks: 50.3 versus 34.0 in the Noometry coding category.
Which has the bigger context window?
GPT-5 does, with 400K tokens against 262K.
How many benchmarks do GPT-5 and Mistral Small share?
28 benchmarks have published results for both models. GPT-5 has 69 scored results on Noometry and Mistral Small has 39.