Model comparison
Mistral Large 3 vs Mistral Medium
Mistral Large 3 is the stronger model overall, scoring 38.6 to 36.0 on the Noometry Index.
Last verified . 20 shared benchmarks.
Mistral AISummary
- They share 20 benchmarks with published results for both. Mistral Large 3 scores higher in 7 categories and Mistral Medium in 2 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Mistral Large 3 leads 35.5 to 24.9.
- The biggest single-benchmark swing is Vectara Hallucination Rate: 14.5% for Mistral Large 3 and 22.7% for Mistral Medium.
- Mistral Large 3 is cheaper at $0.25 / $0.75 per million input/output tokens, against $1.50 / $7.50 for Mistral Medium.
Side by side
| Mistral Large 3 | Mistral Medium | |
|---|---|---|
| Provider | Mistral AI | Mistral AI |
| Noometry Index | 38.6 | 36.0 |
| Released | 2025-12-02 | 2023-12-11 |
| Weights | Open | Open |
| Context window | 262K | 262K |
| Max output | 8K | 262K |
| Input $ / M tokens | $0.25 | $1.50 |
| Output $ / M tokens | $0.75 | $7.50 |
| Results tracked | 24 | 36 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Mistral Large 3: 34.1 (#230), Mistral Medium: 33.8 (#241)
| Benchmark | Mistral Large 3 | Mistral Medium |
|---|---|---|
| LMArena Coding | 1447 | 1434 |
| FrontierCode | — | 8% |
| LMArena WebDev | 1230 | — |
| SciCode | — | 40.2% |
| WeirdML | — | 43.7% |
| ALE-Bench | — | 763.98 |
Agentic & Tool Use Not comparable
Mistral Large 3: —, Mistral Medium: 28.1 (#84)
| Benchmark | Mistral Large 3 | Mistral Medium |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 37.7% |
Reasoning Mistral Medium leads
Mistral Large 3: 14.6 (#314), Mistral Medium: 23.3 (#165)
| Benchmark | Mistral Large 3 | Mistral Medium |
|---|---|---|
| Kagi LLM Benchmark | 50.9% | 50% |
| LMArena Hard Prompts | 1429 | 1426 |
| NYT Connections (extended) | 7.5% | — |
| CritPt | — | 0% |
| Thematic Generalization | 23% | — |
| DTBench | — | 75.5% |
| LMCA | — | 26.1% |
| Surface Evolver Bench | — | 26.9% |
Math Mistral Large 3 leads
Mistral Large 3: 36.0 (#131), Mistral Medium: 27.2 (#238)
| Benchmark | Mistral Large 3 | Mistral Medium |
|---|---|---|
| LMArena Math | 1415 | 1409 |
| OTIS Mock AIME 2024-2025 | — | 32.2% |
| ProofBench | — | 9% |
| MATH Level 5 | — | 81.6% |
| FrontierMath (Feb 2025 set) | — | 0.3% |
Knowledge Mistral Large 3 leads
Mistral Large 3: 35.5 (#173), Mistral Medium: 24.9 (#260)
| Benchmark | Mistral Large 3 | Mistral Medium |
|---|---|---|
| Vectara Hallucination Rate | 14.5% | 22.7% |
| LMArena Expert | 1421 | 1408 |
| GPQA Diamond | — | 59.5% |
| Humanity's Last Exam | — | 4.5% |
Multimodal Mistral Large 3 leads
Mistral Large 3: 37.6 (#65), Mistral Medium: 34.6 (#82)
| Benchmark | Mistral Large 3 | Mistral Medium |
|---|---|---|
| LMArena Vision | 1222 | 1172 |
Multilingual Too close to call
Mistral Large 3: 52.6 (#79), Mistral Medium: 52.3 (#86)
| Benchmark | Mistral Large 3 | Mistral Medium |
|---|---|---|
| LMArena Non-English | 1412 | 1408 |
| LMArena Chinese | 1448 | 1448 |
| LMArena French | 1456 | 1460 |
| LMArena German | 1438 | 1433 |
| LMArena Japanese | 1399 | 1378 |
| LMArena Korean | 1384 | 1382 |
| LMArena Russian | 1412 | 1411 |
| LMArena Spanish | 1437 | 1433 |
Instruction Following Too close to call
Mistral Large 3: 74.1 (#104), Mistral Medium: 73.9 (#111)
| Benchmark | Mistral Large 3 | Mistral Medium |
|---|---|---|
| LMArena Instruction Following | 1402 | 1398 |
Long Context Too close to call
Mistral Large 3: 43.1 (#102), Mistral Medium: 42.9 (#108)
| Benchmark | Mistral Large 3 | Mistral Medium |
|---|---|---|
| LMArena Longer Query | 1413 | 1406 |
Writing & Preference Too close to call
Mistral Large 3: 60.2 (#100), Mistral Medium: 60.2 (#97)
| Benchmark | Mistral Large 3 | Mistral Medium |
|---|---|---|
| LMArena Text | 1427 | 1424 |
| LMArena Creative Writing | 1385 | 1391 |
| LMArena Multi-Turn | 1429 | 1418 |
| Short-Story Creative Writing | — | 77.3% |
| EQ-Bench Creative Writing | 1412 | — |
Frequently asked questions
Is Mistral Large 3 better than Mistral Medium?
Mistral Large 3 is the stronger model overall, scoring 38.6 to 36.0 on the Noometry Index.
Which is cheaper, Mistral Large 3 or Mistral Medium?
Mistral Large 3 is cheaper. It lists at $0.25 per million input tokens and $0.75 per million output tokens; Mistral Medium lists at $1.50 and $7.50.
Is Mistral Large 3 or Mistral Medium better for coding?
They score almost the same on coding (34.1 vs 33.8); test both on your own repository before choosing.
Which has the bigger context window?
Both accept 262K tokens.
How many benchmarks do Mistral Large 3 and Mistral Medium share?
20 benchmarks have published results for both models. Mistral Large 3 has 24 scored results on Noometry and Mistral Medium has 36.