Model comparison
GPT-4 Turbo vs Mistral Medium
Mistral Medium is the stronger model overall, scoring 36.3 to 30.5 on the Noometry Index.
Last verified . 24 shared benchmarks.
Mistral AISummary
- They share 24 benchmarks with published results for both. GPT-4 Turbo scores higher in 0 categories and Mistral Medium in 9 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Mistral Medium leads 28.1 to 9.0.
- The biggest single-benchmark swing is MATH Level 5: 46.7% for GPT-4 Turbo and 81.6% for Mistral Medium.
- Mistral Medium is cheaper at $1.50 / $7.50 per million input/output tokens, against $10 / $30 for GPT-4 Turbo.
- Mistral Medium accepts more context: 262K tokens versus 128K.
- Mistral Medium has downloadable open weights; the other is API-only.
Side by side
| GPT-4 Turbo | Mistral Medium | |
|---|---|---|
| Provider | OpenAI | Mistral AI |
| Noometry Index | 30.5 | 36.3 |
| Released | 2023-11-06 | 2023-12-11 |
| Weights | Proprietary | Open |
| Context window | 128K | 262K |
| Max output | 4K | 262K |
| Input $ / M tokens | $10 | $1.50 |
| Output $ / M tokens | $30 | $7.50 |
| Results tracked | 36 | 36 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
GPT-4 Turbo: 33.8 (#249), Mistral Medium: 34.2 (#243)
| Benchmark | GPT-4 Turbo | Mistral Medium |
|---|---|---|
| WeirdML | 18% | 43.7% |
| LMArena Coding | 1268 | 1434 |
| FrontierCode | — | 8% |
| SciCode | — | 40.2% |
| BigCodeBench Instruct | 48.2% | — |
| BigCodeBench Complete | 58.2% | — |
| ALE-Bench | — | 763.98 |
| HumanEval+ | 86.6% | — |
| MBPP+ | 73.3% | — |
Agentic & Tool Use Not comparable
GPT-4 Turbo: —, Mistral Medium: 28.3 (#90)
| Benchmark | GPT-4 Turbo | Mistral Medium |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 37.7% |
| METR Time Horizons | 36.7% | — |
Reasoning Mistral Medium leads
GPT-4 Turbo: 15.3 (#317), Mistral Medium: 24.0 (#167)
| Benchmark | GPT-4 Turbo | Mistral Medium |
|---|---|---|
| LMArena Hard Prompts | 1251 | 1426 |
| DTBench | 61.6% | 75.5% |
| LMCA | 9.8% | 26.1% |
| SimpleBench | 25.1% | — |
| Kagi LLM Benchmark | — | 50% |
| CritPt | — | 0% |
| Chess Puzzles | 6% | — |
| Surface Evolver Bench | — | 26.9% |
| Epoch Capabilities Index | 127.25 | — |
| ForecastBench | 59.4 | — |
Math Mistral Medium leads
GPT-4 Turbo: 9.0 (#322), Mistral Medium: 28.1 (#245)
| Benchmark | GPT-4 Turbo | Mistral Medium |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 6.7% | 32.2% |
| LMArena Math | 1272 | 1408 |
| MATH Level 5 | 46.7% | 81.6% |
| FrontierMath (Tiers 1-3) | 0.7% | — |
| ProofBench | — | 9% |
| FrontierMath (Feb 2025 set) | — | 0.3% |
Knowledge Too close to call
GPT-4 Turbo: 24.3 (#268), Mistral Medium: 25.0 (#265)
| Benchmark | GPT-4 Turbo | Mistral Medium |
|---|---|---|
| GPQA Diamond | 46.6% | 59.5% |
| LMArena Expert | 1223 | 1408 |
| Humanity's Last Exam | — | 4.5% |
| Confabulations | 28.4% | — |
| Vectara Hallucination Rate | — | 22.7% |
| MMLU | 81.3% | — |
Multimodal Mistral Medium leads
GPT-4 Turbo: 30.6 (#110), Mistral Medium: 35.3 (#88)
| Benchmark | GPT-4 Turbo | Mistral Medium |
|---|---|---|
| LMArena Vision | 1090 | 1172 |
Multilingual Mistral Medium leads
GPT-4 Turbo: 40.5 (#216), Mistral Medium: 52.1 (#91)
| Benchmark | GPT-4 Turbo | Mistral Medium |
|---|---|---|
| LMArena Non-English | 1245 | 1408 |
| LMArena Chinese | 1242 | 1447 |
| LMArena French | 1276 | 1459 |
| LMArena German | 1259 | 1432 |
| LMArena Japanese | 1194 | 1378 |
| LMArena Korean | 1187 | 1380 |
| LMArena Russian | 1259 | 1411 |
| LMArena Spanish | 1260 | 1433 |
Instruction Following Mistral Medium leads
GPT-4 Turbo: 65.8 (#216), Mistral Medium: 73.7 (#116)
| Benchmark | GPT-4 Turbo | Mistral Medium |
|---|---|---|
| LMArena Instruction Following | 1249 | 1398 |
Long Context Mistral Medium leads
GPT-4 Turbo: 38.0 (#206), Mistral Medium: 42.9 (#114)
| Benchmark | GPT-4 Turbo | Mistral Medium |
|---|---|---|
| LMArena Longer Query | 1254 | 1406 |
Writing & Preference Mistral Medium leads
GPT-4 Turbo: 47.7 (#206), Mistral Medium: 60.0 (#103)
| Benchmark | GPT-4 Turbo | Mistral Medium |
|---|---|---|
| LMArena Text | 1272 | 1424 |
| LMArena Creative Writing | 1269 | 1391 |
| LMArena Multi-Turn | 1267 | 1418 |
| Short-Story Creative Writing | — | 77.3% |
Frequently asked questions
Is GPT-4 Turbo better than Mistral Medium?
Mistral Medium is the stronger model overall, scoring 36.3 to 30.5 on the Noometry Index.
Which is cheaper, GPT-4 Turbo 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 Turbo lists at $10 and $30.
Is GPT-4 Turbo or Mistral Medium better for coding?
They score almost the same on coding (33.8 vs 34.2); test both on your own repository before choosing.
Which has the bigger context window?
Mistral Medium does, with 262K tokens against 128K.
How many benchmarks do GPT-4 Turbo and Mistral Medium share?
24 benchmarks have published results for both models. GPT-4 Turbo has 36 scored results on Noometry and Mistral Medium has 36.