Model comparison
GPT-4 vs Mistral
GPT-4 and Mistral score almost the same on the Noometry Index (29.1 vs 29.9), so choose on price, context window or the category you care about most.
Last verified . 17 shared benchmarks.
Mistral AISummary
- They share 17 benchmarks with published results for both. GPT-4 scores higher in 4 categories and Mistral in 4 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in instruction following, where GPT-4 leads 65.3 to 52.6.
Side by side
| GPT-4 | Mistral | |
|---|---|---|
| Provider | OpenAI | Mistral AI |
| Noometry Index | 29.1 | 29.9 |
| Released | 2023-03-14 | — |
| Weights | Proprietary | Proprietary |
| Context window | 8K | — |
| Max output | 8K | — |
| Input $ / M tokens | $30 | — |
| Output $ / M tokens | $60 | — |
| Results tracked | 38 | 22 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Mistral leads
GPT-4: 31.6 (#283), Mistral: 33.8 (#250)
| Benchmark | GPT-4 | Mistral |
|---|---|---|
| LMArena Coding | 1254 | 1162 |
| WeirdML | 12.4% | — |
| BigCodeBench Instruct | 46% | — |
| BigCodeBench Complete | 57.2% | — |
| HumanEval+ | 79.3% | — |
Agentic & Tool Use Not comparable
GPT-4: —, Mistral: —
| Benchmark | GPT-4 | Mistral |
|---|---|---|
| METR Time Horizons | 36.1% | — |
Reasoning Mistral leads
GPT-4: 17.8 (#289), Mistral: 22.2 (#200)
| Benchmark | GPT-4 | Mistral |
|---|---|---|
| LMArena Hard Prompts | 1241 | 1149 |
| Chess Puzzles | 4% | — |
| Mystery Game Puzzles | 12% | — |
| DTBench | 62.7% | — |
| LMCA | 17.1% | — |
| BIG-Bench Hard | 75.1% | — |
| Epoch Capabilities Index | 125.89 | — |
| ForecastBench | 57.8 | — |
| HellaSwag | 95.3% | — |
| WinoGrande | 87.5% | — |
Math Mistral leads
GPT-4: 10.8 (#309), Mistral: 22.3 (#278)
| Benchmark | GPT-4 | Mistral |
|---|---|---|
| LMArena Math | 1269 | 1180 |
| OTIS Mock AIME 2024-2025 | 1.1% | — |
| Omni-MATH | — | 7.2% |
| MATH Level 5 | 23% | — |
| GSM8K | 92% | — |
Knowledge GPT-4 leads
GPT-4: 18.4 (#282), Mistral: 16.6 (#288)
| Benchmark | GPT-4 | Mistral |
|---|---|---|
| LMArena Expert | 1211 | 1125 |
| GPQA Diamond | 35.7% | — |
| MMLU-Pro | — | 27.7% |
| GPQA (HELM) | — | 30.3% |
| MMLU | 86.4% | — |
| TriviaQA | 84.8% | — |
Multilingual GPT-4 leads
GPT-4: 40.6 (#215), Mistral: 32.8 (#254)
| Benchmark | GPT-4 | Mistral |
|---|---|---|
| LMArena Non-English | 1246 | 1129 |
| LMArena Chinese | 1242 | 1109 |
| LMArena French | 1283 | 1180 |
| LMArena German | 1251 | 1155 |
| LMArena Japanese | 1209 | 1013 |
| LMArena Korean | 1184 | 1032 |
| LMArena Russian | 1251 | 1168 |
| LMArena Spanish | 1261 | 1143 |
Instruction Following GPT-4 leads
GPT-4: 65.3 (#222), Mistral: 52.6 (#288)
| Benchmark | GPT-4 | Mistral |
|---|---|---|
| LMArena Instruction Following | 1241 | 1152 |
| IFEval | — | 56.8% |
Long Context GPT-4 leads
GPT-4: 37.7 (#212), Mistral: 35.0 (#245)
| Benchmark | GPT-4 | Mistral |
|---|---|---|
| LMArena Longer Query | 1244 | 1153 |
Writing & Preference Mistral leads
GPT-4: 34.9 (#268), Mistral: 37.0 (#260)
| Benchmark | GPT-4 | Mistral |
|---|---|---|
| LMArena Text | 1263 | 1165 |
| LMArena Creative Writing | 1244 | 1158 |
| LMArena Multi-Turn | 1257 | 1147 |
| EQ-Bench Creative Writing | 752 | — |
| WildBench | — | 66% |
Frequently asked questions
Is GPT-4 better than Mistral?
GPT-4 and Mistral score almost the same on the Noometry Index (29.1 vs 29.9), so choose on price, context window or the category you care about most.
Is GPT-4 or Mistral better for coding?
Mistral scores higher on coding benchmarks: 33.8 versus 31.6 in the Noometry coding category.
How many benchmarks do GPT-4 and Mistral share?
17 benchmarks have published results for both models. GPT-4 has 38 scored results on Noometry and Mistral has 22.