Model comparison
Mistral Large vs Mistral Large 3
Mistral Large 3 is the stronger model overall, scoring 39.1 to 31.9 on the Noometry Index.
Last verified . 19 shared benchmarks.
Mistral AISummary
- They share 19 benchmarks with published results for both. Mistral Large scores higher in 1 category and Mistral Large 3 in 7 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in math, where Mistral Large 3 leads 38.7 to 18.2.
- The biggest single-benchmark swing is Vectara Hallucination Rate: 4.5% for Mistral Large and 14.5% for Mistral Large 3.
- Mistral Large 3 is cheaper at $0.25 / $0.75 per million input/output tokens, against $2 / $6 for Mistral Large.
- Mistral Large 3 accepts more context: 262K tokens versus 131K.
Side by side
| Mistral Large | Mistral Large 3 | |
|---|---|---|
| Provider | Mistral AI | Mistral AI |
| Noometry Index | 31.9 | 39.1 |
| Released | 2024-02-26 | 2025-12-02 |
| Weights | Open | Open |
| Context window | 131K | 262K |
| Max output | 16K | 8K |
| Input $ / M tokens | $2 | $0.25 |
| Output $ / M tokens | $6 | $0.75 |
| Results tracked | 51 | 24 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Mistral Large: 34.3 (#240), Mistral Large 3: 34.4 (#237)
| Benchmark | Mistral Large | Mistral Large 3 |
|---|---|---|
| LMArena Coding | 1277 | 1448 |
| LMArena WebDev | — | 1230 |
| SciCode | 36.2% | — |
| BigCodeBench Instruct | 30% | — |
| LiveBench Coding | 47.1% | — |
| BigCodeBench Complete | 38.3% | — |
| ALE-Bench | 264.7 | — |
| HumanEval+ | 62.2% | — |
| MBPP+ | 59.5% | — |
Agentic & Tool Use Not comparable
Mistral Large: 28.6 (#89), Mistral Large 3: —
| Benchmark | Mistral Large | Mistral Large 3 |
|---|---|---|
| Berkeley Function Calling Leaderboard | 38.4% | — |
Reasoning Too close to call
Mistral Large: 15.8 (#310), Mistral Large 3: 15.2 (#319)
| Benchmark | Mistral Large | Mistral Large 3 |
|---|---|---|
| LMArena Hard Prompts | 1257 | 1429 |
| SimpleBench | 22.5% | — |
| Kagi LLM Benchmark | — | 50.9% |
| NYT Connections (extended) | — | 7.5% |
| CritPt | 0% | — |
| Thematic Generalization | — | 23% |
| LiveBench Reasoning | 43.5% | — |
| DTBench | 65.1% | — |
| LiveBench Data Analysis | 50.1% | — |
| LMCA | 16.7% | — |
| Epoch Capabilities Index | 128.52 | — |
| ForecastBench | 57.1 | — |
| LiveBench | 48.4% | — |
Math Mistral Large 3 leads
Mistral Large: 18.2 (#291), Mistral Large 3: 38.7 (#129)
| Benchmark | Mistral Large | Mistral Large 3 |
|---|---|---|
| LMArena Math | 1262 | 1414 |
| OTIS Mock AIME 2024-2025 | 8.5% | — |
| Omni-MATH | 28.1% | — |
| LiveBench Math | 42.5% | — |
| MATH Level 5 | 50.3% | — |
| FrontierMath (Feb 2025 set) | 0.3% | — |
Knowledge Mistral Large 3 leads
Mistral Large: 30.1 (#230), Mistral Large 3: 36.0 (#177)
| Benchmark | Mistral Large | Mistral Large 3 |
|---|---|---|
| Vectara Hallucination Rate | 4.5% | 14.5% |
| LMArena Expert | 1232 | 1421 |
| GPQA Diamond | 51.3% | — |
| MMLU-Pro | 59.9% | — |
| Confabulations | 21.4% | — |
| GPQA (HELM) | 43.5% | — |
| MMLU | 80% | — |
Multimodal Not comparable
Mistral Large: —, Mistral Large 3: 38.2 (#66)
| Benchmark | Mistral Large | Mistral Large 3 |
|---|---|---|
| LMArena Vision | — | 1221 |
Multilingual Mistral Large 3 leads
Mistral Large: 40.0 (#219), Mistral Large 3: 52.5 (#84)
| Benchmark | Mistral Large | Mistral Large 3 |
|---|---|---|
| LMArena Non-English | 1237 | 1413 |
| LMArena Chinese | 1240 | 1447 |
| LMArena French | 1325 | 1455 |
| LMArena German | 1254 | 1437 |
| LMArena Japanese | 1188 | 1394 |
| LMArena Korean | 1202 | 1384 |
| LMArena Russian | 1257 | 1411 |
| LMArena Spanish | 1268 | 1440 |
Instruction Following Mistral Large 3 leads
Mistral Large: 67.9 (#191), Mistral Large 3: 74.0 (#108)
| Benchmark | Mistral Large | Mistral Large 3 |
|---|---|---|
| LMArena Instruction Following | 1249 | 1403 |
| LiveBench Instruction Following | 67.9% | — |
| IFEval | 87.7% | — |
Long Context Mistral Large 3 leads
Mistral Large: 38.3 (#199), Mistral Large 3: 43.1 (#105)
| Benchmark | Mistral Large | Mistral Large 3 |
|---|---|---|
| LMArena Longer Query | 1261 | 1413 |
Writing & Preference Mistral Large 3 leads
Mistral Large: 40.7 (#242), Mistral Large 3: 60.0 (#101)
| Benchmark | Mistral Large | Mistral Large 3 |
|---|---|---|
| LMArena Text | 1266 | 1428 |
| LMArena Creative Writing | 1243 | 1386 |
| EQ-Bench Creative Writing | 985 | 1412 |
| LMArena Multi-Turn | 1260 | 1429 |
| Short-Story Creative Writing | 69% | — |
| WildBench | 80.1% | — |
| LiveBench Language | 39.4% | — |
Frequently asked questions
Is Mistral Large better than Mistral Large 3?
Mistral Large 3 is the stronger model overall, scoring 39.1 to 31.9 on the Noometry Index.
Which is cheaper, Mistral Large or Mistral Large 3?
Mistral Large 3 is cheaper. It lists at $0.25 per million input tokens and $0.75 per million output tokens; Mistral Large lists at $2 and $6.
Is Mistral Large or Mistral Large 3 better for coding?
They score almost the same on coding (34.3 vs 34.4); test both on your own repository before choosing.
Which has the bigger context window?
Mistral Large 3 does, with 262K tokens against 131K.
How many benchmarks do Mistral Large and Mistral Large 3 share?
19 benchmarks have published results for both models. Mistral Large has 51 scored results on Noometry and Mistral Large 3 has 24.