Model comparison
DeepSeek-V3.2-Exp vs Mistral Large
DeepSeek-V3.2-Exp is the stronger model overall, scoring 44.3 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-V3.2-Exp 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-V3.2-Exp leads 41.7 to 18.2.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 87.8% for DeepSeek-V3.2-Exp and 8.5% for Mistral Large.
- DeepSeek-V3.2-Exp is cheaper at $0.26 / $0.38 per million input/output tokens, against $2 / $6 for Mistral Large.
- DeepSeek-V3.2-Exp accepts more context: 164K tokens versus 131K.
Side by side
| DeepSeek-V3.2-Exp | Mistral Large | |
|---|---|---|
| Provider | DeepSeek | Mistral AI |
| Noometry Index | 44.3 | 31.9 |
| Released | 2025-09-29 | 2024-02-26 |
| Weights | Open | Open |
| Context window | 164K | 131K |
| Max output | 66K | 16K |
| Input $ / M tokens | $0.26 | $2 |
| Output $ / M tokens | $0.38 | $6 |
| Results tracked | 49 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 46.5 (#65), Mistral Large: 34.3 (#240)
| Benchmark | DeepSeek-V3.2-Exp | Mistral Large |
|---|---|---|
| SciCode | 38.9% | 36.2% |
| LMArena Coding | 1454 | 1277 |
| SWE-bench Verified (bash only) | 70% | — |
| Aider Polyglot | 74.2% | — |
| LMArena WebDev | 1362 | — |
| SWE-bench Multilingual | 59% | — |
| WeirdML | 39.5% | — |
| BigCodeBench Instruct | — | 30% |
| LiveBench Coding | — | 47.1% |
| BigCodeBench Complete | — | 38.3% |
| ALE-Bench | — | 264.7 |
| HumanEval+ | — | 62.2% |
| MBPP+ | — | 59.5% |
Agentic & Tool Use DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 32.7 (#59), Mistral Large: 28.6 (#89)
| Benchmark | DeepSeek-V3.2-Exp | Mistral Large |
|---|---|---|
| Berkeley Function Calling Leaderboard | 56.7% | 38.4% |
| Terminal-Bench | 39.6% | — |
| APEX-Agents | 21.3% | — |
| TheAgentCompany | 42.9% | — |
| Vending-Bench 2 | 1,034 | — |
Reasoning DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 22.1 (#208), Mistral Large: 15.8 (#310)
| Benchmark | DeepSeek-V3.2-Exp | Mistral Large |
|---|---|---|
| CritPt | 2.9% | 0% |
| LMArena Hard Prompts | 1434 | 1257 |
| DTBench | 87.7% | 65.1% |
| LMCA | 29.1% | 16.7% |
| Epoch Capabilities Index | 146.27 | 128.52 |
| ARC-AGI-2 | 4% | — |
| SimpleBench | — | 22.5% |
| Kagi LLM Benchmark | 52.2% | — |
| NYT Connections (extended) | 36.7% | — |
| ARC-AGI-1 | 57% | — |
| Chess Puzzles | 14% | — |
| Thematic Generalization | 65% | — |
| LiveBench Reasoning | — | 43.5% |
| LiveBench Data Analysis | — | 50.1% |
| ForecastBench | — | 57.1 |
| LiveBench | — | 48.4% |
Math DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 41.7 (#87), Mistral Large: 18.2 (#291)
| Benchmark | DeepSeek-V3.2-Exp | Mistral Large |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 87.8% | 8.5% |
| LMArena Math | 1435 | 1262 |
| FrontierMath (Feb 2025 set) | 22.1% | 0.3% |
| MathArena Final-Answer Competitions | 57.7% | — |
| ProofBench | 8% | — |
| Omni-MATH | — | 28.1% |
| LiveBench Math | — | 42.5% |
| MATH Level 5 | — | 50.3% |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 51.7 (#66), Mistral Large: 30.1 (#230)
| Benchmark | DeepSeek-V3.2-Exp | Mistral Large |
|---|---|---|
| GPQA Diamond | 83.4% | 51.3% |
| Vectara Hallucination Rate | 5.3% | 4.5% |
| LMArena Expert | 1436 | 1232 |
| MMLU-Pro | — | 59.9% |
| Confabulations | — | 21.4% |
| GPQA (HELM) | — | 43.5% |
| MMLU | — | 80% |
Multilingual DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 52.2 (#90), Mistral Large: 40.0 (#219)
| Benchmark | DeepSeek-V3.2-Exp | Mistral Large |
|---|---|---|
| LMArena Non-English | 1409 | 1237 |
| LMArena Chinese | 1461 | 1240 |
| LMArena French | 1433 | 1325 |
| LMArena German | 1440 | 1254 |
| LMArena Japanese | 1374 | 1188 |
| LMArena Korean | 1371 | 1202 |
| LMArena Russian | 1424 | 1257 |
| LMArena Spanish | 1440 | 1268 |
Instruction Following DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 74.5 (#93), Mistral Large: 67.9 (#191)
| Benchmark | DeepSeek-V3.2-Exp | Mistral Large |
|---|---|---|
| LMArena Instruction Following | 1413 | 1249 |
| LiveBench Instruction Following | — | 67.9% |
| IFEval | — | 87.7% |
Long Context DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 47.6 (#16), Mistral Large: 38.3 (#199)
| Benchmark | DeepSeek-V3.2-Exp | Mistral Large |
|---|---|---|
| LMArena Longer Query | 1428 | 1261 |
| Fiction.LiveBench | 83.3% | — |
| CL-bench | 13.2% | — |
| CL-bench Life | 9.5% | — |
Writing & Preference DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 62.4 (#77), Mistral Large: 40.7 (#242)
| Benchmark | DeepSeek-V3.2-Exp | Mistral Large |
|---|---|---|
| LMArena Text | 1425 | 1266 |
| LMArena Creative Writing | 1403 | 1243 |
| EQ-Bench Creative Writing | 1515 | 985 |
| LMArena Multi-Turn | 1427 | 1260 |
| Short-Story Creative Writing | — | 69% |
| WildBench | — | 80.1% |
| LiveBench Language | — | 39.4% |
Frequently asked questions
Is DeepSeek-V3.2-Exp better than Mistral Large?
DeepSeek-V3.2-Exp is the stronger model overall, scoring 44.3 to 31.9 on the Noometry Index.
Which is cheaper, DeepSeek-V3.2-Exp or Mistral Large?
DeepSeek-V3.2-Exp is cheaper. It lists at $0.26 per million input tokens and $0.38 per million output tokens; Mistral Large lists at $2 and $6.
Is DeepSeek-V3.2-Exp or Mistral Large better for coding?
DeepSeek-V3.2-Exp scores higher on coding benchmarks: 46.5 versus 34.3 in the Noometry coding category.
Which has the bigger context window?
DeepSeek-V3.2-Exp does, with 164K tokens against 131K.
How many benchmarks do DeepSeek-V3.2-Exp and Mistral Large share?
28 benchmarks have published results for both models. DeepSeek-V3.2-Exp has 49 scored results on Noometry and Mistral Large has 51.