Model comparison
GPT-4 vs Mistral Medium
Mistral Medium is the stronger model overall, scoring 36.3 to 29.1 on the Noometry Index.
Last verified . 23 shared benchmarks.
Mistral AISummary
- They share 23 benchmarks with published results for both. GPT-4 scores higher in 0 categories and Mistral Medium in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Mistral Medium leads 60.0 to 34.9.
- The biggest single-benchmark swing is MATH Level 5: 23% for GPT-4 and 81.6% for Mistral Medium.
- Mistral Medium is cheaper at $1.50 / $7.50 per million input/output tokens, against $30 / $60 for GPT-4.
- Mistral Medium accepts more context: 262K tokens versus 8K.
- Mistral Medium has downloadable open weights; the other is API-only.
Side by side
| GPT-4 | Mistral Medium | |
|---|---|---|
| Provider | OpenAI | Mistral AI |
| Noometry Index | 29.1 | 36.3 |
| Released | 2023-03-14 | 2023-12-11 |
| Weights | Proprietary | Open |
| Context window | 8K | 262K |
| Max output | 8K | 262K |
| Input $ / M tokens | $30 | $1.50 |
| Output $ / M tokens | $60 | $7.50 |
| Results tracked | 38 | 36 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Mistral Medium leads
GPT-4: 31.6 (#283), Mistral Medium: 34.2 (#243)
| Benchmark | GPT-4 | Mistral Medium |
|---|---|---|
| WeirdML | 12.4% | 43.7% |
| LMArena Coding | 1254 | 1434 |
| FrontierCode | — | 8% |
| SciCode | — | 40.2% |
| BigCodeBench Instruct | 46% | — |
| BigCodeBench Complete | 57.2% | — |
| ALE-Bench | — | 763.98 |
| HumanEval+ | 79.3% | — |
Agentic & Tool Use Not comparable
GPT-4: —, Mistral Medium: 28.3 (#90)
| Benchmark | GPT-4 | Mistral Medium |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 37.7% |
| METR Time Horizons | 36.1% | — |
Reasoning Mistral Medium leads
GPT-4: 17.8 (#289), Mistral Medium: 24.0 (#167)
| Benchmark | GPT-4 | Mistral Medium |
|---|---|---|
| LMArena Hard Prompts | 1241 | 1426 |
| DTBench | 62.7% | 75.5% |
| LMCA | 17.1% | 26.1% |
| Kagi LLM Benchmark | — | 50% |
| CritPt | — | 0% |
| Chess Puzzles | 4% | — |
| Mystery Game Puzzles | 12% | — |
| Surface Evolver Bench | — | 26.9% |
| BIG-Bench Hard | 75.1% | — |
| Epoch Capabilities Index | 125.89 | — |
| ForecastBench | 57.8 | — |
| HellaSwag | 95.3% | — |
| WinoGrande | 87.5% | — |
Math Mistral Medium leads
GPT-4: 10.8 (#309), Mistral Medium: 28.1 (#245)
| Benchmark | GPT-4 | Mistral Medium |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.1% | 32.2% |
| LMArena Math | 1269 | 1408 |
| MATH Level 5 | 23% | 81.6% |
| ProofBench | — | 9% |
| FrontierMath (Feb 2025 set) | — | 0.3% |
| GSM8K | 92% | — |
Knowledge Mistral Medium leads
GPT-4: 18.4 (#282), Mistral Medium: 25.0 (#265)
| Benchmark | GPT-4 | Mistral Medium |
|---|---|---|
| GPQA Diamond | 35.7% | 59.5% |
| LMArena Expert | 1211 | 1408 |
| Humanity's Last Exam | — | 4.5% |
| Vectara Hallucination Rate | — | 22.7% |
| MMLU | 86.4% | — |
| TriviaQA | 84.8% | — |
Multimodal Not comparable
GPT-4: —, Mistral Medium: 35.3 (#88)
| Benchmark | GPT-4 | Mistral Medium |
|---|---|---|
| LMArena Vision | — | 1172 |
Multilingual Mistral Medium leads
GPT-4: 40.6 (#215), Mistral Medium: 52.1 (#91)
| Benchmark | GPT-4 | Mistral Medium |
|---|---|---|
| LMArena Non-English | 1246 | 1408 |
| LMArena Chinese | 1242 | 1447 |
| LMArena French | 1283 | 1459 |
| LMArena German | 1251 | 1432 |
| LMArena Japanese | 1209 | 1378 |
| LMArena Korean | 1184 | 1380 |
| LMArena Russian | 1251 | 1411 |
| LMArena Spanish | 1261 | 1433 |
Instruction Following Mistral Medium leads
GPT-4: 65.3 (#222), Mistral Medium: 73.7 (#116)
| Benchmark | GPT-4 | Mistral Medium |
|---|---|---|
| LMArena Instruction Following | 1241 | 1398 |
Long Context Mistral Medium leads
GPT-4: 37.7 (#212), Mistral Medium: 42.9 (#114)
| Benchmark | GPT-4 | Mistral Medium |
|---|---|---|
| LMArena Longer Query | 1244 | 1406 |
Writing & Preference Mistral Medium leads
GPT-4: 34.9 (#268), Mistral Medium: 60.0 (#103)
| Benchmark | GPT-4 | Mistral Medium |
|---|---|---|
| LMArena Text | 1263 | 1424 |
| LMArena Creative Writing | 1244 | 1391 |
| LMArena Multi-Turn | 1257 | 1418 |
| Short-Story Creative Writing | — | 77.3% |
| EQ-Bench Creative Writing | 752 | — |
Frequently asked questions
Is GPT-4 better than Mistral Medium?
Mistral Medium is the stronger model overall, scoring 36.3 to 29.1 on the Noometry Index.
Which is cheaper, GPT-4 or Mistral Medium?
Mistral Medium is cheaper. It lists at $1.50 per million input tokens and $7.50 per million output tokens; GPT-4 lists at $30 and $60.
Is GPT-4 or Mistral Medium better for coding?
Mistral Medium scores higher on coding benchmarks: 34.2 versus 31.6 in the Noometry coding category.
Which has the bigger context window?
Mistral Medium does, with 262K tokens against 8K.
How many benchmarks do GPT-4 and Mistral Medium share?
23 benchmarks have published results for both models. GPT-4 has 38 scored results on Noometry and Mistral Medium has 36.