Model comparison
Mistral Large 3 vs o3
o3 is the stronger model overall, scoring 47.5 to 39.1 on the Noometry Index. Mistral Large 3 costs 9.3× less per token, which makes it the better buy when o3's lead doesn't matter for your workload.
Last verified . 20 shared benchmarks.
Mistral AISummary
- They share 20 benchmarks with published results for both. Mistral Large 3 scores higher in 2 categories and o3 in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where o3 leads 54.6 to 36.0.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 50.9% for Mistral Large 3 and 67.6% for o3.
- Mistral Large 3 is cheaper at $0.25 / $0.75 per million input/output tokens, against $2 / $8 for o3.
- Mistral Large 3 accepts more context: 262K tokens versus 200K.
- Mistral Large 3 has downloadable open weights; the other is API-only.
Side by side
| Mistral Large 3 | o3 | |
|---|---|---|
| Provider | Mistral AI | OpenAI |
| Noometry Index | 39.1 | 47.5 |
| Released | 2025-12-02 | 2025-04-16 |
| Weights | Open | Proprietary |
| Context window | 262K | 200K |
| Max output | 8K | 100K |
| Input $ / M tokens | $0.25 | $2 |
| Output $ / M tokens | $0.75 | $8 |
| Results tracked | 24 | 63 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o3 leads
Mistral Large 3: 34.4 (#237), o3: 46.8 (#64)
| Benchmark | Mistral Large 3 | o3 |
|---|---|---|
| LMArena Coding | 1448 | 1408 |
| SWE-bench Verified | — | 62.3% |
| SWE-bench Verified (bash only) | — | 58.4% |
| Aider Polyglot | — | 81.3% |
| LMArena WebDev | 1230 | — |
| GSO | — | 8.8% |
| WeirdML | — | 52.4% |
| CadEval | — | 74% |
| ALE-Bench | — | 933.55 |
Agentic & Tool Use Not comparable
Mistral Large 3: —, o3: 34.5 (#44)
| Benchmark | Mistral Large 3 | o3 |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 63% |
| GDPval | — | 30.8% |
| DeepResearch Bench | — | 45.2% |
| OSWorld | — | 23% |
| LMArena Search | — | 1144 |
| METR Time Horizons | — | 65.4% |
Reasoning o3 leads
Mistral Large 3: 15.2 (#319), o3: 32.0 (#78)
| Benchmark | Mistral Large 3 | o3 |
|---|---|---|
| Kagi LLM Benchmark | 50.9% | 67.6% |
| LMArena Hard Prompts | 1429 | 1402 |
| ARC-AGI-2 | — | 6.5% |
| SimpleBench | — | 53.1% |
| NYT Connections (extended) | 7.5% | — |
| ARC-AGI-1 | — | 60.8% |
| CritPt | — | 1.4% |
| Chess Puzzles | — | 38% |
| EnigmaEval | — | 13.1% |
| Thematic Generalization | 23% | — |
| Mystery Game Puzzles | — | 29% |
| DTBench | — | 84.8% |
| LMCA | — | 39.7% |
| Epoch Capabilities Index | — | 146.86 |
| ForecastBench | — | 62.5 |
Math o3 leads
Mistral Large 3: 38.7 (#129), o3: 50.2 (#58)
| Benchmark | Mistral Large 3 | o3 |
|---|---|---|
| LMArena Math | 1414 | 1426 |
| FrontierMath (Tiers 1-3) | — | 33.3% |
| OTIS Mock AIME 2024-2025 | — | 84.4% |
| Omni-MATH | — | 71.4% |
| MATH Level 5 | — | 97.8% |
| FrontierMath (Feb 2025 set) | — | 18.7% |
| FrontierMath Tier 4 (v1) | — | 2.1% |
Knowledge o3 leads
Mistral Large 3: 36.0 (#177), o3: 54.6 (#52)
| Benchmark | Mistral Large 3 | o3 |
|---|---|---|
| LMArena Expert | 1421 | 1402 |
| GPQA Diamond | — | 81.8% |
| Humanity's Last Exam | — | 20.3% |
| SimpleQA Verified | — | 49.4% |
| MMLU-Pro | — | 85.9% |
| Confabulations | — | 14.4% |
| Vectara Hallucination Rate | 14.5% | — |
| GPQA (HELM) | — | 75.3% |
Multimodal o3 leads
Mistral Large 3: 38.2 (#66), o3: 41.4 (#36)
| Benchmark | Mistral Large 3 | o3 |
|---|---|---|
| LMArena Vision | 1221 | 1214 |
| GeoBench | — | 74% |
| VPCT | — | 52% |
Multilingual Too close to call
Mistral Large 3: 52.5 (#84), o3: 51.7 (#105)
| Benchmark | Mistral Large 3 | o3 |
|---|---|---|
| LMArena Non-English | 1413 | 1401 |
| LMArena Chinese | 1447 | 1437 |
| LMArena French | 1455 | 1430 |
| LMArena German | 1437 | 1420 |
| LMArena Japanese | 1394 | 1403 |
| LMArena Korean | 1384 | 1370 |
| LMArena Russian | 1411 | 1406 |
| LMArena Spanish | 1440 | 1395 |
Instruction Following Mistral Large 3 leads
Mistral Large 3: 74.0 (#108), o3: 72.8 (#127)
| Benchmark | Mistral Large 3 | o3 |
|---|---|---|
| LMArena Instruction Following | 1403 | 1368 |
| IFEval | — | 86.9% |
Long Context o3 leads
Mistral Large 3: 43.1 (#105), o3: 53.3 (#6)
| Benchmark | Mistral Large 3 | o3 |
|---|---|---|
| LMArena Longer Query | 1413 | 1372 |
| Fiction.LiveBench | — | 88.9% |
| CL-bench | — | 17.8% |
Writing & Preference o3 leads
Mistral Large 3: 60.0 (#101), o3: 63.5 (#64)
| Benchmark | Mistral Large 3 | o3 |
|---|---|---|
| LMArena Text | 1428 | 1410 |
| LMArena Creative Writing | 1386 | 1359 |
| EQ-Bench Creative Writing | 1412 | 1676 |
| LMArena Multi-Turn | 1429 | 1405 |
| Short-Story Creative Writing | — | 83.9% |
| WildBench | — | 86.1% |
Frequently asked questions
Is Mistral Large 3 better than o3?
o3 is the stronger model overall, scoring 47.5 to 39.1 on the Noometry Index. Mistral Large 3 costs 9.3× less per token, which makes it the better buy when o3's lead doesn't matter for your workload.
Which is cheaper, Mistral Large 3 or o3?
Mistral Large 3 is cheaper. It lists at $0.25 per million input tokens and $0.75 per million output tokens; o3 lists at $2 and $8.
Is Mistral Large 3 or o3 better for coding?
o3 scores higher on coding benchmarks: 46.8 versus 34.4 in the Noometry coding category.
Which has the bigger context window?
Mistral Large 3 does, with 262K tokens against 200K.
How many benchmarks do Mistral Large 3 and o3 share?
20 benchmarks have published results for both models. Mistral Large 3 has 24 scored results on Noometry and o3 has 63.