Model comparison
DeepSeek-V3 vs Mistral Large 4
Mistral Large 4 is the stronger model overall, scoring 43.1 to 39.5 on the Noometry Index. DeepSeek-V3 costs 2.5× less per token, which makes it the better buy when Mistral Large 4's lead doesn't matter for your workload.
Last verified . 12 shared benchmarks.
DeepSeek
Mistral AISummary
- They share 12 benchmarks with published results for both. DeepSeek-V3 scores higher in 1 category and Mistral Large 4 in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in long context, where Mistral Large 4 leads 43.6 to 34.0.
- DeepSeek-V3 is cheaper at $0.24 / $0.90 per million input/output tokens, against $0.68 / $2.09 for Mistral Large 4.
- Mistral Large 4 accepts more context: 1.05M tokens versus 164K.
- DeepSeek-V3 has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V3 | Mistral Large 4 | |
|---|---|---|
| Provider | DeepSeek | Mistral AI |
| Noometry Index | 39.5 | 43.1 |
| Released | 2024-12-26 | 2026-10-06 |
| Weights | Open | Proprietary |
| Context window | 164K | 1.05M |
| Max output | 164K | 262K |
| Input $ / M tokens | $0.24 | $0.68 |
| Output $ / M tokens | $0.90 | $2.09 |
| Results tracked | 60 | 15 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Mistral Large 4 leads
DeepSeek-V3: 42.3 (#106), Mistral Large 4: 48.6 (#57)
| Benchmark | DeepSeek-V3 | Mistral Large 4 |
|---|---|---|
| LMArena Coding | 1368 | 1475 |
| Aider Polyglot | 55.1% | — |
| LMArena WebDev | — | 1541 |
| SciCode | 35.8% | — |
| WeirdML | 36.1% | — |
| BigCodeBench Instruct | 50% | — |
| LiveBench Coding | 70.9% | — |
| BigCodeBench Complete | 62.2% | — |
| HumanEval+ | 86.6% | — |
| MBPP+ | 73% | — |
Agentic & Tool Use Not comparable
DeepSeek-V3: —, Mistral Large 4: —
| Benchmark | DeepSeek-V3 | Mistral Large 4 |
|---|---|---|
| METR Time Horizons | 49.6% | — |
Reasoning Mistral Large 4 leads
DeepSeek-V3: 20.5 (#236), Mistral Large 4: 22.5 (#192)
| Benchmark | DeepSeek-V3 | Mistral Large 4 |
|---|---|---|
| LMArena Hard Prompts | 1365 | 1444 |
| SimpleBench | 27.2% | — |
| Kagi LLM Benchmark | 52.3% | — |
| NYT Connections (extended) | — | 27.4% |
| CritPt | 0% | — |
| LiveBench Reasoning | 65.8% | — |
| DTBench | 64.8% | — |
| LiveBench Data Analysis | 60.9% | — |
| LMCA | 15.5% | — |
| BIG-Bench Hard | 87.5% | — |
| Epoch Capabilities Index | 135.94 | — |
| ForecastBench | 59.1 | — |
| HellaSwag | 88.9% | — |
| LiveBench | 66.9% | — |
| PIQA | 84.7% | — |
| WinoGrande | 85.2% | — |
Math Mistral Large 4 leads
DeepSeek-V3: 32.1 (#219), Mistral Large 4: 40.4 (#91)
| Benchmark | DeepSeek-V3 | Mistral Large 4 |
|---|---|---|
| LMArena Math | 1373 | 1488 |
| OTIS Mock AIME 2024-2025 | 37.8% | — |
| Omni-MATH | 40.3% | — |
| LiveBench Math | 73.5% | — |
| MATH Level 5 | 75.5% | — |
| FrontierMath (Feb 2025 set) | 1.7% | — |
Knowledge Too close to call
DeepSeek-V3: 37.5 (#155), Mistral Large 4: 36.6 (#166)
| Benchmark | DeepSeek-V3 | Mistral Large 4 |
|---|---|---|
| LMArena Expert | 1351 | 1447 |
| GPQA Diamond | 67.6% | — |
| SimpleQA Verified | — | 20% |
| MMLU-Pro | 72.3% | — |
| Confabulations | 26.1% | — |
| Vectara Hallucination Rate | 6.1% | — |
| GPQA (HELM) | 53.8% | — |
| ARC (AI2) Challenge | 95.3% | — |
| MMLU | 87.2% | — |
| TriviaQA | 82.9% | — |
Multilingual Mistral Large 4 leads
DeepSeek-V3: 48.5 (#143), Mistral Large 4: 52.6 (#82)
| Benchmark | DeepSeek-V3 | Mistral Large 4 |
|---|---|---|
| LMArena Non-English | 1358 | 1415 |
| LMArena Chinese | 1391 | 1491 |
| LMArena Russian | 1373 | 1414 |
| LMArena French | 1385 | — |
| LMArena German | 1374 | — |
| LMArena Japanese | 1333 | — |
| LMArena Korean | 1319 | — |
| LMArena Spanish | 1358 | — |
Instruction Following Mistral Large 4 leads
DeepSeek-V3: 72.8 (#130), Mistral Large 4: 75.0 (#76)
| Benchmark | DeepSeek-V3 | Mistral Large 4 |
|---|---|---|
| LMArena Instruction Following | 1345 | 1424 |
| LiveBench Instruction Following | 81.5% | — |
| IFEval | 83.2% | — |
Long Context Mistral Large 4 leads
DeepSeek-V3: 34.0 (#253), Mistral Large 4: 43.6 (#89)
| Benchmark | DeepSeek-V3 | Mistral Large 4 |
|---|---|---|
| LMArena Longer Query | 1352 | 1429 |
| Fiction.LiveBench | 50% | — |
Writing & Preference Mistral Large 4 leads
DeepSeek-V3: 57.4 (#130), Mistral Large 4: 60.4 (#97)
| Benchmark | DeepSeek-V3 | Mistral Large 4 |
|---|---|---|
| LMArena Text | 1375 | 1427 |
| LMArena Creative Writing | 1364 | 1361 |
| LMArena Multi-Turn | 1389 | 1424 |
| Short-Story Creative Writing | 77% | — |
| EQ-Bench Creative Writing | 1472 | — |
| WildBench | 83% | — |
| LiveBench Language | 49.1% | — |
Frequently asked questions
Is DeepSeek-V3 better than Mistral Large 4?
Mistral Large 4 is the stronger model overall, scoring 43.1 to 39.5 on the Noometry Index. DeepSeek-V3 costs 2.5× less per token, which makes it the better buy when Mistral Large 4's lead doesn't matter for your workload.
Which is cheaper, DeepSeek-V3 or Mistral Large 4?
DeepSeek-V3 is cheaper. It lists at $0.24 per million input tokens and $0.90 per million output tokens; Mistral Large 4 lists at $0.68 and $2.09.
Is DeepSeek-V3 or Mistral Large 4 better for coding?
Mistral Large 4 scores higher on coding benchmarks: 48.6 versus 42.3 in the Noometry coding category.
Which has the bigger context window?
Mistral Large 4 does, with 1.05M tokens against 164K.
How many benchmarks do DeepSeek-V3 and Mistral Large 4 share?
12 benchmarks have published results for both models. DeepSeek-V3 has 60 scored results on Noometry and Mistral Large 4 has 15.