Model comparison
GPT-4o mini vs Mistral Small
Mistral Small is the stronger model overall, scoring 33.1 to 25.4 on the Noometry Index.
Last verified . 34 shared benchmarks.
Mistral AISummary
- They share 34 benchmarks with published results for both. GPT-4o mini scores higher in 0 categories and Mistral Small in 10 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Mistral Small leads 30.9 to 17.7.
- The biggest single-benchmark swing is DTBench: 54.4% for GPT-4o mini and 70.9% for Mistral Small.
- Both cost about the same: $0.15 input and $0.60 output per million tokens.
- Mistral Small accepts more context: 256K tokens versus 128K.
- Mistral Small has downloadable open weights; the other is API-only.
Side by side
| GPT-4o mini | Mistral Small | |
|---|---|---|
| Provider | OpenAI | Mistral AI |
| Noometry Index | 25.4 | 33.1 |
| Released | 2024-07-18 | 2024-02-26 |
| Weights | Proprietary | Open |
| Context window | 128K | 256K |
| Max output | 16K | 256K |
| Input $ / M tokens | $0.15 | $0.15 |
| Output $ / M tokens | $0.60 | $0.60 |
| Results tracked | 60 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Mistral Small leads
GPT-4o mini: 21.6 (#330), Mistral Small: 33.5 (#242)
| Benchmark | GPT-4o mini | Mistral Small |
|---|---|---|
| BigCodeBench Instruct | 46.1% | 36.1% |
| LiveBench Coding | 43.1% | 36.2% |
| LMArena Coding | 1290 | 1363 |
| BigCodeBench Complete | 57.4% | 46.6% |
| Aider Polyglot | 3.6% | — |
| SciCode | — | 26.5% |
| WeirdML | 11.8% | — |
| ALE-Bench | — | 497.62 |
| HumanEval+ | 83.5% | — |
| MBPP+ | 72.2% | — |
Agentic & Tool Use Too close to call
GPT-4o mini: 27.2 (#95), Mistral Small: 27.9 (#87)
| Benchmark | GPT-4o mini | Mistral Small |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 37.1% |
| BALROG | 17.4% | — |
Reasoning Mistral Small leads
GPT-4o mini: 8.7 (#343), Mistral Small: 19.3 (#238)
| Benchmark | GPT-4o mini | Mistral Small |
|---|---|---|
| Kagi LLM Benchmark | 28.8% | 37.8% |
| LiveBench Reasoning | 32.8% | 44.8% |
| LMArena Hard Prompts | 1267 | 1336 |
| DTBench | 54.4% | 70.9% |
| LiveBench Data Analysis | 50% | 53.7% |
| LMCA | 10.4% | 20.6% |
| LiveBench | 41.3% | 44% |
| ARC-AGI-2 | 0% | — |
| SimpleBench | 10.7% | — |
| CritPt | — | 0% |
| Chess Puzzles | 0% | — |
| Mystery Game Puzzles | 12% | — |
| Epoch Capabilities Index | 126.56 | — |
| PIQA | 88.7% | — |
Math Mistral Small leads
GPT-4o mini: 9.9 (#307), Mistral Small: 15.4 (#287)
| Benchmark | GPT-4o mini | Mistral Small |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 6.9% | 5.8% |
| LiveBench Math | 36.3% | 39.9% |
| LMArena Math | 1267 | 1341 |
| MATH Level 5 | 52.6% | 46.8% |
| FrontierMath (Tiers 1-3) | 0.7% | — |
| Omni-MATH | 28% | — |
| GSM8K | 91.3% | — |
Knowledge Mistral Small leads
GPT-4o mini: 17.7 (#279), Mistral Small: 30.9 (#214)
| Benchmark | GPT-4o mini | Mistral Small |
|---|---|---|
| GPQA Diamond | 37.7% | 47.5% |
| LMArena Expert | 1235 | 1293 |
| MMLU | 81.8% | 68.7% |
| SimpleQA Verified | 8.3% | — |
| MMLU-Pro | 60.3% | — |
| Confabulations | 37.2% | — |
| Vectara Hallucination Rate | — | 5.1% |
| GPQA (HELM) | 36.8% | — |
| BoolQ | 88.7% | — |
Multimodal Mistral Small leads
GPT-4o mini: 25.3 (#117), Mistral Small: 32.9 (#91)
| Benchmark | GPT-4o mini | Mistral Small |
|---|---|---|
| LMArena Vision | 1066 | 1142 |
| Video-MME | 64.8% | — |
| GeoBench | 64% | — |
| VPCT | 34% | — |
Multilingual Mistral Small leads
GPT-4o mini: 42.2 (#194), Mistral Small: 45.7 (#164)
| Benchmark | GPT-4o mini | Mistral Small |
|---|---|---|
| LMArena Non-English | 1266 | 1315 |
| LMArena Chinese | 1264 | 1342 |
| LMArena French | 1296 | 1337 |
| LMArena German | 1272 | 1341 |
| LMArena Japanese | 1216 | 1275 |
| LMArena Korean | 1195 | 1260 |
| LMArena Russian | 1275 | 1324 |
| LMArena Spanish | 1276 | 1346 |
Instruction Following Mistral Small leads
GPT-4o mini: 62.1 (#234), Mistral Small: 66.6 (#204)
| Benchmark | GPT-4o mini | Mistral Small |
|---|---|---|
| LiveBench Instruction Following | 56.8% | 63.7% |
| LMArena Instruction Following | 1259 | 1310 |
| IFEval | 78.2% | — |
Long Context Mistral Small leads
GPT-4o mini: 39.1 (#181), Mistral Small: 40.4 (#151)
| Benchmark | GPT-4o mini | Mistral Small |
|---|---|---|
| LMArena Longer Query | 1289 | 1327 |
Writing & Preference Mistral Small leads
GPT-4o mini: 39.7 (#243), Mistral Small: 52.6 (#165)
| Benchmark | GPT-4o mini | Mistral Small |
|---|---|---|
| LMArena Text | 1286 | 1338 |
| LMArena Creative Writing | 1268 | 1304 |
| LMArena Multi-Turn | 1285 | 1344 |
| LiveBench Language | 28.6% | 30.5% |
| Short-Story Creative Writing | 67.2% | — |
| EQ-Bench Creative Writing | 873 | — |
| WildBench | 79.1% | — |
Frequently asked questions
Is GPT-4o mini better than Mistral Small?
Mistral Small is the stronger model overall, scoring 33.1 to 25.4 on the Noometry Index.
Which is cheaper, GPT-4o mini or Mistral Small?
Mistral Small is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; GPT-4o mini lists at $0.15 and $0.60.
Is GPT-4o mini or Mistral Small better for coding?
Mistral Small scores higher on coding benchmarks: 33.5 versus 21.6 in the Noometry coding category.
Which has the bigger context window?
Mistral Small does, with 256K tokens against 128K.
How many benchmarks do GPT-4o mini and Mistral Small share?
34 benchmarks have published results for both models. GPT-4o mini has 60 scored results on Noometry and Mistral Small has 39.