Model comparison
Mistral Large vs o1
o1 is the stronger model overall, scoring 40.9 to 31.9 on the Noometry Index. Mistral Large costs 8.8× less per token, which makes it the better buy when o1's lead doesn't matter for your workload.
Last verified . 36 shared benchmarks.
Mistral AISummary
- They share 36 benchmarks with published results for both. Mistral Large scores higher in 1 category and o1 in 8 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where o1 leads 36.1 to 18.2.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 8.5% for Mistral Large and 73.3% for o1.
- Mistral Large is cheaper at $2 / $6 per million input/output tokens, against $15 / $60 for o1.
- o1 accepts more context: 200K tokens versus 131K.
- Mistral Large has downloadable open weights; the other is API-only.
Side by side
| Mistral Large | o1 | |
|---|---|---|
| Provider | Mistral AI | OpenAI |
| Noometry Index | 31.9 | 40.9 |
| Released | 2024-02-26 | 2024-09-12 |
| Weights | Open | Proprietary |
| Context window | 131K | 200K |
| Max output | 16K | 100K |
| Input $ / M tokens | $2 | $15 |
| Output $ / M tokens | $6 | $60 |
| Results tracked | 51 | 52 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o1 leads
Mistral Large: 34.3 (#240), o1: 46.1 (#70)
| Benchmark | Mistral Large | o1 |
|---|---|---|
| LiveBench Coding | 47.1% | 69.7% |
| LMArena Coding | 1277 | 1367 |
| HumanEval+ | 62.2% | 89% |
| MBPP+ | 59.5% | 80.2% |
| Aider Polyglot | — | 61.7% |
| SciCode | 36.2% | — |
| WeirdML | — | 47.6% |
| BigCodeBench Instruct | 30% | — |
| BigCodeBench Complete | 38.3% | — |
| CadEval | — | 56% |
| ALE-Bench | 264.7 | — |
Agentic & Tool Use Mistral Large leads
Mistral Large: 28.6 (#89), o1: 24.6 (#117)
| Benchmark | Mistral Large | o1 |
|---|---|---|
| Berkeley Function Calling Leaderboard | 38.4% | — |
| Cybench | — | 10% |
| METR Time Horizons | — | 51.1% |
Reasoning o1 leads
Mistral Large: 15.8 (#310), o1: 27.9 (#111)
| Benchmark | Mistral Large | o1 |
|---|---|---|
| SimpleBench | 22.5% | 41.7% |
| LiveBench Reasoning | 43.5% | 91.6% |
| LMArena Hard Prompts | 1257 | 1371 |
| DTBench | 65.1% | 74.7% |
| LiveBench Data Analysis | 50.1% | 65.5% |
| LMCA | 16.7% | 22.3% |
| Epoch Capabilities Index | 128.52 | 141.91 |
| LiveBench | 48.4% | 75.7% |
| ARC-AGI-1 | — | 30.7% |
| CritPt | 0% | — |
| Chess Puzzles | — | 15% |
| EnigmaEval | — | 5.7% |
| ForecastBench | 57.1 | — |
Math o1 leads
Mistral Large: 18.2 (#291), o1: 36.1 (#175)
| Benchmark | Mistral Large | o1 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 8.5% | 73.3% |
| LiveBench Math | 42.5% | 80.3% |
| LMArena Math | 1262 | 1388 |
| MATH Level 5 | 50.3% | 94.7% |
| FrontierMath (Feb 2025 set) | 0.3% | 9.3% |
| FrontierMath (Tiers 1-3) | — | 14.7% |
| Omni-MATH | 28.1% | — |
Knowledge o1 leads
Mistral Large: 30.1 (#230), o1: 41.5 (#110)
| Benchmark | Mistral Large | o1 |
|---|---|---|
| GPQA Diamond | 51.3% | 76.8% |
| Confabulations | 21.4% | 11.7% |
| LMArena Expert | 1232 | 1361 |
| Humanity's Last Exam | — | 8% |
| SimpleQA Verified | — | 41.1% |
| MMLU-Pro | 59.9% | — |
| Vectara Hallucination Rate | 4.5% | — |
| GPQA (HELM) | 43.5% | — |
| MMLU | 80% | — |
Multimodal Not comparable
Mistral Large: —, o1: 34.2 (#93)
| Benchmark | Mistral Large | o1 |
|---|---|---|
| LMArena Vision | — | 1168 |
| GeoBench | — | 80% |
| VPCT | — | 37% |
| SpatialViz-Bench | — | 41.4% |
Multilingual o1 leads
Mistral Large: 40.0 (#219), o1: 48.6 (#142)
| Benchmark | Mistral Large | o1 |
|---|---|---|
| LMArena Non-English | 1237 | 1358 |
| LMArena Chinese | 1240 | 1394 |
| LMArena French | 1325 | 1344 |
| LMArena German | 1254 | 1337 |
| LMArena Japanese | 1188 | 1346 |
| LMArena Korean | 1202 | 1396 |
| LMArena Russian | 1257 | 1356 |
| LMArena Spanish | 1268 | 1345 |
Instruction Following o1 leads
Mistral Large: 67.9 (#191), o1: 74.8 (#86)
| Benchmark | Mistral Large | o1 |
|---|---|---|
| LiveBench Instruction Following | 67.9% | 81.5% |
| LMArena Instruction Following | 1249 | 1367 |
| IFEval | 87.7% | — |
Long Context o1 leads
Mistral Large: 38.3 (#199), o1: 50.3 (#9)
| Benchmark | Mistral Large | o1 |
|---|---|---|
| LMArena Longer Query | 1261 | 1378 |
| Fiction.LiveBench | — | 83.3% |
Writing & Preference o1 leads
Mistral Large: 40.7 (#242), o1: 55.6 (#144)
| Benchmark | Mistral Large | o1 |
|---|---|---|
| LMArena Text | 1266 | 1366 |
| LMArena Creative Writing | 1243 | 1348 |
| Short-Story Creative Writing | 69% | 70.2% |
| LMArena Multi-Turn | 1260 | 1369 |
| LiveBench Language | 39.4% | 65.4% |
| EQ-Bench Creative Writing | 985 | — |
| WildBench | 80.1% | — |
Frequently asked questions
Is Mistral Large better than o1?
o1 is the stronger model overall, scoring 40.9 to 31.9 on the Noometry Index. Mistral Large costs 8.8× less per token, which makes it the better buy when o1's lead doesn't matter for your workload.
Which is cheaper, Mistral Large or o1?
Mistral Large is cheaper. It lists at $2 per million input tokens and $6 per million output tokens; o1 lists at $15 and $60.
Is Mistral Large or o1 better for coding?
o1 scores higher on coding benchmarks: 46.1 versus 34.3 in the Noometry coding category.
Which has the bigger context window?
o1 does, with 200K tokens against 131K.
How many benchmarks do Mistral Large and o1 share?
36 benchmarks have published results for both models. Mistral Large has 51 scored results on Noometry and o1 has 52.