Model comparison
GPT-4o mini vs Mistral Medium
Mistral Medium is the stronger model overall, scoring 36.3 to 25.5 on the Noometry Index. GPT-4o mini costs 11× less per token, which makes it the better buy when Mistral Medium's lead doesn't matter for your workload.
Last verified . 26 shared benchmarks.
Mistral AISummary
- They share 26 benchmarks with published results for both. GPT-4o mini scores higher in 0 categories and Mistral Medium in 10 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Mistral Medium leads 60.0 to 39.5.
- The biggest single-benchmark swing is WeirdML: 11.8% for GPT-4o mini and 43.7% for Mistral Medium.
- GPT-4o mini is cheaper at $0.15 / $0.60 per million input/output tokens, against $1.50 / $7.50 for Mistral Medium.
- Mistral Medium accepts more context: 262K tokens versus 128K.
- Mistral Medium has downloadable open weights; the other is API-only.
Side by side
| GPT-4o mini | Mistral Medium | |
|---|---|---|
| Provider | OpenAI | Mistral AI |
| Noometry Index | 25.5 | 36.3 |
| Released | 2024-07-18 | 2023-12-11 |
| Weights | Proprietary | Open |
| Context window | 128K | 262K |
| Max output | 16K | 262K |
| Input $ / M tokens | $0.15 | $1.50 |
| Output $ / M tokens | $0.60 | $7.50 |
| Results tracked | 60 | 36 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Mistral Medium leads
GPT-4o mini: 22.0 (#335), Mistral Medium: 34.2 (#243)
| Benchmark | GPT-4o mini | Mistral Medium |
|---|---|---|
| WeirdML | 11.8% | 43.7% |
| LMArena Coding | 1290 | 1434 |
| FrontierCode | — | 8% |
| Aider Polyglot | 3.6% | — |
| SciCode | — | 40.2% |
| BigCodeBench Instruct | 46.1% | — |
| LiveBench Coding | 43.1% | — |
| BigCodeBench Complete | 57.4% | — |
| ALE-Bench | — | 763.98 |
| HumanEval+ | 83.5% | — |
| MBPP+ | 72.2% | — |
Agentic & Tool Use Too close to call
GPT-4o mini: 27.5 (#101), Mistral Medium: 28.3 (#90)
| Benchmark | GPT-4o mini | Mistral Medium |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 37.7% |
| BALROG | 17.4% | — |
Reasoning Mistral Medium leads
GPT-4o mini: 8.7 (#347), Mistral Medium: 24.0 (#167)
| Benchmark | GPT-4o mini | Mistral Medium |
|---|---|---|
| Kagi LLM Benchmark | 28.8% | 50% |
| LMArena Hard Prompts | 1267 | 1426 |
| DTBench | 54.4% | 75.5% |
| LMCA | 10.4% | 26.1% |
| ARC-AGI-2 | 0% | — |
| SimpleBench | 10.7% | — |
| CritPt | — | 0% |
| Chess Puzzles | 0% | — |
| LiveBench Reasoning | 32.8% | — |
| Mystery Game Puzzles | 12% | — |
| LiveBench Data Analysis | 50% | — |
| Surface Evolver Bench | — | 26.9% |
| Epoch Capabilities Index | 126.56 | — |
| LiveBench | 41.3% | — |
| PIQA | 88.7% | — |
Math Mistral Medium leads
GPT-4o mini: 10.4 (#314), Mistral Medium: 28.1 (#245)
| Benchmark | GPT-4o mini | Mistral Medium |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 6.9% | 32.2% |
| LMArena Math | 1267 | 1408 |
| MATH Level 5 | 52.6% | 81.6% |
| FrontierMath (Tiers 1-3) | 0.7% | — |
| ProofBench | — | 9% |
| Omni-MATH | 28% | — |
| LiveBench Math | 36.3% | — |
| FrontierMath (Feb 2025 set) | — | 0.3% |
| GSM8K | 91.3% | — |
Knowledge Mistral Medium leads
GPT-4o mini: 17.7 (#284), Mistral Medium: 25.0 (#265)
| Benchmark | GPT-4o mini | Mistral Medium |
|---|---|---|
| GPQA Diamond | 37.7% | 59.5% |
| LMArena Expert | 1235 | 1408 |
| Humanity's Last Exam | — | 4.5% |
| SimpleQA Verified | 8.3% | — |
| MMLU-Pro | 60.3% | — |
| Confabulations | 37.2% | — |
| Vectara Hallucination Rate | — | 22.7% |
| GPQA (HELM) | 36.8% | — |
| BoolQ | 88.7% | — |
| MMLU | 81.8% | — |
Multimodal Mistral Medium leads
GPT-4o mini: 25.9 (#122), Mistral Medium: 35.3 (#88)
| Benchmark | GPT-4o mini | Mistral Medium |
|---|---|---|
| LMArena Vision | 1066 | 1172 |
| Video-MME | 64.8% | — |
| GeoBench | 64% | — |
| VPCT | 34% | — |
Multilingual Mistral Medium leads
GPT-4o mini: 42.0 (#199), Mistral Medium: 52.1 (#91)
| Benchmark | GPT-4o mini | Mistral Medium |
|---|---|---|
| LMArena Non-English | 1266 | 1408 |
| LMArena Chinese | 1265 | 1447 |
| LMArena French | 1297 | 1459 |
| LMArena German | 1272 | 1432 |
| LMArena Japanese | 1216 | 1378 |
| LMArena Korean | 1195 | 1380 |
| LMArena Russian | 1275 | 1411 |
| LMArena Spanish | 1276 | 1433 |
Instruction Following Mistral Medium leads
GPT-4o mini: 61.9 (#239), Mistral Medium: 73.7 (#116)
| Benchmark | GPT-4o mini | Mistral Medium |
|---|---|---|
| LMArena Instruction Following | 1258 | 1398 |
| LiveBench Instruction Following | 56.8% | — |
| IFEval | 78.2% | — |
Long Context Mistral Medium leads
GPT-4o mini: 39.1 (#186), Mistral Medium: 42.9 (#114)
| Benchmark | GPT-4o mini | Mistral Medium |
|---|---|---|
| LMArena Longer Query | 1289 | 1406 |
Writing & Preference Mistral Medium leads
GPT-4o mini: 39.5 (#248), Mistral Medium: 60.0 (#103)
| Benchmark | GPT-4o mini | Mistral Medium |
|---|---|---|
| LMArena Text | 1286 | 1424 |
| LMArena Creative Writing | 1268 | 1391 |
| Short-Story Creative Writing | 67.2% | 77.3% |
| LMArena Multi-Turn | 1285 | 1418 |
| EQ-Bench Creative Writing | 873 | — |
| WildBench | 79.1% | — |
| LiveBench Language | 28.6% | — |
Frequently asked questions
Is GPT-4o mini better than Mistral Medium?
Mistral Medium is the stronger model overall, scoring 36.3 to 25.5 on the Noometry Index. GPT-4o mini costs 11× less per token, which makes it the better buy when Mistral Medium's lead doesn't matter for your workload.
Which is cheaper, GPT-4o mini or Mistral Medium?
GPT-4o mini is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; Mistral Medium lists at $1.50 and $7.50.
Is GPT-4o mini or Mistral Medium better for coding?
Mistral Medium scores higher on coding benchmarks: 34.2 versus 22.0 in the Noometry coding category.
Which has the bigger context window?
Mistral Medium does, with 262K tokens against 128K.
How many benchmarks do GPT-4o mini and Mistral Medium share?
26 benchmarks have published results for both models. GPT-4o mini has 60 scored results on Noometry and Mistral Medium has 36.