Model comparison
DeepSeek V4 Pro vs Mistral Large
DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 31.9 on the Noometry Index.
Last verified . 28 shared benchmarks.
DeepSeek
Mistral AISummary
- They share 28 benchmarks with published results for both. DeepSeek V4 Pro scores higher in 9 categories and Mistral Large in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek V4 Pro leads 64.8 to 18.2.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 98.6% for DeepSeek V4 Pro and 8.5% for Mistral Large.
- DeepSeek V4 Pro is cheaper at $0.66 / $1.98 per million input/output tokens, against $2 / $6 for Mistral Large.
- DeepSeek V4 Pro accepts more context: 1M tokens versus 131K.
Side by side
| DeepSeek V4 Pro | Mistral Large | |
|---|---|---|
| Provider | DeepSeek | Mistral AI |
| Noometry Index | 54.3 | 31.9 |
| Released | 2026-04-24 | 2024-02-26 |
| Weights | Open | Open |
| Context window | 1M | 131K |
| Max output | 393K | 16K |
| Input $ / M tokens | $0.66 | $2 |
| Output $ / M tokens | $1.98 | $6 |
| Results tracked | 48 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek V4 Pro leads
DeepSeek V4 Pro: 52.4 (#34), Mistral Large: 34.3 (#240)
| Benchmark | DeepSeek V4 Pro | Mistral Large |
|---|---|---|
| SciCode | 51% | 36.2% |
| LMArena Coding | 1470 | 1277 |
| ALE-Bench | 1,403 | 264.7 |
| SWE-bench Verified | 77.6% | — |
| FrontierCode | 28.6% | — |
| LMArena WebDev | 1582 | — |
| WeirdML | 66.2% | — |
| BigCodeBench Instruct | — | 30% |
| LiveBench Coding | — | 47.1% |
| BigCodeBench Complete | — | 38.3% |
| HumanEval+ | — | 62.2% |
| MBPP+ | — | 59.5% |
Agentic & Tool Use DeepSeek V4 Pro leads
DeepSeek V4 Pro: 32.8 (#58), Mistral Large: 28.6 (#89)
| Benchmark | DeepSeek V4 Pro | Mistral Large |
|---|---|---|
| APEX-Agents | 47.3% | — |
| Berkeley Function Calling Leaderboard | — | 38.4% |
| Vending-Bench 2 | 3,285 | — |
Reasoning DeepSeek V4 Pro leads
DeepSeek V4 Pro: 56.5 (#24), Mistral Large: 15.8 (#310)
| Benchmark | DeepSeek V4 Pro | Mistral Large |
|---|---|---|
| CritPt | 18% | 0% |
| LMArena Hard Prompts | 1461 | 1257 |
| DTBench | 93.9% | 65.1% |
| LMCA | 45.5% | 16.7% |
| Epoch Capabilities Index | 155.31 | 128.52 |
| ForecastBench | 56.1 | 57.1 |
| ARC-AGI-2 | 61.3% | — |
| SimpleBench | — | 22.5% |
| Kagi LLM Benchmark | 53.5% | — |
| NYT Connections (extended) | 91.3% | — |
| ARC-AGI-1 | 90.5% | — |
| Chess Puzzles | 47% | — |
| LiveBench Reasoning | — | 43.5% |
| Mystery Game Puzzles | 43% | — |
| LiveBench Data Analysis | — | 50.1% |
| Surface Evolver Bench | 40% | — |
| LiveBench | — | 48.4% |
Math DeepSeek V4 Pro leads
DeepSeek V4 Pro: 64.8 (#30), Mistral Large: 18.2 (#291)
| Benchmark | DeepSeek V4 Pro | Mistral Large |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 98.6% | 8.5% |
| LMArena Math | 1455 | 1262 |
| FrontierMath (Tiers 1-3) | 64.6% | — |
| FrontierMath Tier 4 | 26.8% | — |
| MathArena Final-Answer Competitions | 76.6% | — |
| ProofBench | 50% | — |
| Omni-MATH | — | 28.1% |
| LiveBench Math | — | 42.5% |
| MATH Level 5 | — | 50.3% |
| FrontierMath (Feb 2025 set) | — | 0.3% |
Knowledge DeepSeek V4 Pro leads
DeepSeek V4 Pro: 59.5 (#31), Mistral Large: 30.1 (#230)
| Benchmark | DeepSeek V4 Pro | Mistral Large |
|---|---|---|
| GPQA Diamond | 91.7% | 51.3% |
| Vectara Hallucination Rate | 8.6% | 4.5% |
| LMArena Expert | 1464 | 1232 |
| SimpleQA Verified | 52.9% | — |
| MMLU-Pro | — | 59.9% |
| Confabulations | — | 21.4% |
| GPQA (HELM) | — | 43.5% |
| MMLU | — | 80% |
Multilingual DeepSeek V4 Pro leads
DeepSeek V4 Pro: 54.4 (#45), Mistral Large: 40.0 (#219)
| Benchmark | DeepSeek V4 Pro | Mistral Large |
|---|---|---|
| LMArena Non-English | 1439 | 1237 |
| LMArena Chinese | 1486 | 1240 |
| LMArena French | 1472 | 1325 |
| LMArena German | 1458 | 1254 |
| LMArena Japanese | 1445 | 1188 |
| LMArena Korean | 1447 | 1202 |
| LMArena Russian | 1453 | 1257 |
| LMArena Spanish | 1458 | 1268 |
Instruction Following DeepSeek V4 Pro leads
DeepSeek V4 Pro: 76.1 (#47), Mistral Large: 67.9 (#191)
| Benchmark | DeepSeek V4 Pro | Mistral Large |
|---|---|---|
| LMArena Instruction Following | 1448 | 1249 |
| LiveBench Instruction Following | — | 67.9% |
| IFEval | — | 87.7% |
Long Context DeepSeek V4 Pro leads
DeepSeek V4 Pro: 45.0 (#51), Mistral Large: 38.3 (#199)
| Benchmark | DeepSeek V4 Pro | Mistral Large |
|---|---|---|
| LMArena Longer Query | 1458 | 1261 |
| CL-bench Life | 13.5% | — |
Writing & Preference DeepSeek V4 Pro leads
DeepSeek V4 Pro: 65.5 (#46), Mistral Large: 40.7 (#242)
| Benchmark | DeepSeek V4 Pro | Mistral Large |
|---|---|---|
| LMArena Text | 1451 | 1266 |
| LMArena Creative Writing | 1446 | 1243 |
| EQ-Bench Creative Writing | 1553 | 985 |
| LMArena Multi-Turn | 1467 | 1260 |
| Short-Story Creative Writing | — | 69% |
| WildBench | — | 80.1% |
| EQ-Bench 4 | 1166 | — |
| LiveBench Language | — | 39.4% |
Frequently asked questions
Is DeepSeek V4 Pro better than Mistral Large?
DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 31.9 on the Noometry Index.
Which is cheaper, DeepSeek V4 Pro or Mistral Large?
DeepSeek V4 Pro is cheaper. It lists at $0.66 per million input tokens and $1.98 per million output tokens; Mistral Large lists at $2 and $6.
Is DeepSeek V4 Pro or Mistral Large better for coding?
DeepSeek V4 Pro scores higher on coding benchmarks: 52.4 versus 34.3 in the Noometry coding category.
Which has the bigger context window?
DeepSeek V4 Pro does, with 1M tokens against 131K.
How many benchmarks do DeepSeek V4 Pro and Mistral Large share?
28 benchmarks have published results for both models. DeepSeek V4 Pro has 48 scored results on Noometry and Mistral Large has 51.