# DeepSeek-V3.2-Speciale vs Mistral Large

> DeepSeek-V3.2-Speciale is the stronger model overall, scoring 39.1 to 31.6 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-v3-2-speciale-vs-mistral-large
- Last updated: 2026-10-10
- Shared benchmarks: 2

## Summary

- They share 2 benchmarks with published results for both. DeepSeek-V3.2-Speciale scores higher in 3 categories and Mistral Large in 0 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek-V3.2-Speciale leads 32.1 to 15.4.
- The biggest single-benchmark swing is SimpleBench: 52.6% for DeepSeek-V3.2-Speciale and 22.5% for Mistral Large.
- DeepSeek-V3.2-Speciale is cheaper at $0.58 / $1.68 per million input/output tokens, against $2 / $6 for Mistral Large.
- Mistral Large accepts more context: 131K tokens versus 128K.

## Snapshot

| | DeepSeek-V3.2-Speciale | Mistral Large |
|---|---|---|
| Provider | DeepSeek | Mistral AI |
| Noometry Index | 39.1 | 31.6 |
| Rank | 159 | 258 |
| Context | 128K | 131K |
| Input $/M | $0.58 | $2 |
| Output $/M | $1.68 | $6 |
| Weights | Open | Open |

## Coding

- DeepSeek-V3.2-Speciale: 39.9 (#136)
- Mistral Large: 33.8 (#240)

| Benchmark | DeepSeek-V3.2-Speciale | Mistral Large |
|---|---|---|
| SciCode | — | 36.2% |
| WeirdML | 46.7% | — |
| BigCodeBench Instruct | — | 30% |
| LiveBench Coding | — | 47.1% |
| LMArena Coding | — | 1277 |
| BigCodeBench Complete | — | 38.3% |
| ALE-Bench | — | 264.7 |
| HumanEval+ | — | 62.2% |
| MBPP+ | — | 59.5% |

## Agentic & Tool Use

- DeepSeek-V3.2-Speciale: —
- Mistral Large: 28.4 (#83)

| Benchmark | DeepSeek-V3.2-Speciale | Mistral Large |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 38.4% |

## Reasoning

- DeepSeek-V3.2-Speciale: 32.1 (#65)
- Mistral Large: 15.4 (#305)

| Benchmark | DeepSeek-V3.2-Speciale | Mistral Large |
|---|---|---|
| SimpleBench | 52.6% | 22.5% |
| CritPt | — | 0% |
| LiveBench Reasoning | — | 43.5% |
| LMArena Hard Prompts | — | 1257 |
| DTBench | — | 65.1% |
| LiveBench Data Analysis | — | 50.1% |
| LMCA | — | 16.7% |
| Epoch Capabilities Index | — | 128.52 |
| ForecastBench | — | 57.1 |
| LiveBench | — | 48.4% |

## Math

- DeepSeek-V3.2-Speciale: —
- Mistral Large: 17.1 (#285)

| Benchmark | DeepSeek-V3.2-Speciale | Mistral Large |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 8.5% |
| Omni-MATH | — | 28.1% |
| LiveBench Math | — | 42.5% |
| LMArena Math | — | 1262 |
| MATH Level 5 | — | 50.3% |
| FrontierMath (Feb 2025 set) | — | 0.3% |

## Knowledge

- DeepSeek-V3.2-Speciale: —
- Mistral Large: 30.0 (#224)

| Benchmark | DeepSeek-V3.2-Speciale | Mistral Large |
|---|---|---|
| GPQA Diamond | — | 51.3% |
| MMLU-Pro | — | 59.9% |
| Confabulations | — | 21.4% |
| Vectara Hallucination Rate | — | 4.5% |
| GPQA (HELM) | — | 43.5% |
| LMArena Expert | — | 1232 |
| MMLU | — | 80% |

## Multilingual

- DeepSeek-V3.2-Speciale: —
- Mistral Large: 40.1 (#214)

| Benchmark | DeepSeek-V3.2-Speciale | Mistral Large |
|---|---|---|
| LMArena Non-English | — | 1237 |
| LMArena Chinese | — | 1240 |
| LMArena French | — | 1324 |
| LMArena German | — | 1254 |
| LMArena Japanese | — | 1187 |
| LMArena Korean | — | 1202 |
| LMArena Russian | — | 1257 |
| LMArena Spanish | — | 1268 |

## Instruction Following

- DeepSeek-V3.2-Speciale: —
- Mistral Large: 68.1 (#186)

| Benchmark | DeepSeek-V3.2-Speciale | Mistral Large |
|---|---|---|
| LiveBench Instruction Following | — | 67.9% |
| IFEval | — | 87.7% |
| LMArena Instruction Following | — | 1249 |

## Long Context

- DeepSeek-V3.2-Speciale: —
- Mistral Large: 38.3 (#194)

| Benchmark | DeepSeek-V3.2-Speciale | Mistral Large |
|---|---|---|
| LMArena Longer Query | — | 1262 |

## Writing & Preference

- DeepSeek-V3.2-Speciale: 46.1 (#217)
- Mistral Large: 40.9 (#237)

| Benchmark | DeepSeek-V3.2-Speciale | Mistral Large |
|---|---|---|
| EQ-Bench Creative Writing | 1276 | 985 |
| LMArena Text | — | 1266 |
| LMArena Creative Writing | — | 1243 |
| Short-Story Creative Writing | — | 69% |
| WildBench | — | 80.1% |
| LMArena Multi-Turn | — | 1260 |
| LiveBench Language | — | 39.4% |

## FAQ

### Is DeepSeek-V3.2-Speciale better than Mistral Large?

DeepSeek-V3.2-Speciale is the stronger model overall, scoring 39.1 to 31.6 on the Noometry Index.

### Which is cheaper, DeepSeek-V3.2-Speciale or Mistral Large?

DeepSeek-V3.2-Speciale is cheaper. It lists at $0.58 per million input tokens and $1.68 per million output tokens; Mistral Large lists at $2 and $6.

### Is DeepSeek-V3.2-Speciale or Mistral Large better for coding?

DeepSeek-V3.2-Speciale scores higher on coding benchmarks: 39.9 versus 33.8 in the Noometry coding category.

### Which has the bigger context window?

Mistral Large does, with 131K tokens against 128K.

### How many benchmarks do DeepSeek-V3.2-Speciale and Mistral Large share?

2 benchmarks have published results for both models. DeepSeek-V3.2-Speciale has 3 scored results on Noometry and Mistral Large has 51.