# DeepSeek-R1 vs Mistral Medium 3.5

> DeepSeek-R1 is the stronger model overall, scoring 42.3 to 40.2 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-r1-vs-mistral-medium-3-5
- Last updated: 2026-10-10
- Shared benchmarks: 18

## Summary

- They share 18 benchmarks with published results for both. DeepSeek-R1 scores higher in 7 categories and Mistral Medium 3.5 in 1 category; 7 gaps are clear of the uncertainty.
- The widest gap is in coding, where DeepSeek-R1 leads 46.3 to 36.0.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 69.4% for DeepSeek-R1 and 41.4% for Mistral Medium 3.5.
- DeepSeek-R1 is cheaper at $0.50 / $2.15 per million input/output tokens, against $1.50 / $7.50 for Mistral Medium 3.5.
- Mistral Medium 3.5 accepts more context: 262K tokens versus 164K.
- Mistral Medium 3.5 has downloadable open weights; the other is API-only.

## Snapshot

| | DeepSeek-R1 | Mistral Medium 3.5 |
|---|---|---|
| Provider | DeepSeek | Mistral AI |
| Noometry Index | 42.3 | 40.2 |
| Rank | 115 | 152 |
| Context | 164K | 262K |
| Input $/M | $0.50 | $1.50 |
| Output $/M | $2.15 | $7.50 |
| Weights | Proprietary | Open |

## Coding

- DeepSeek-R1: 46.3 (#68)
- Mistral Medium 3.5: 36.0 (#213)

| Benchmark | DeepSeek-R1 | Mistral Medium 3.5 |
|---|---|---|
| LMArena Coding | 1427 | 1461 |
| Aider Polyglot | 71.4% | — |
| LMArena WebDev | — | 1264 |
| SciCode | 35.7% | — |
| WeirdML | 41.6% | — |
| LiveBench Coding | 66.7% | — |
| ALE-Bench | 804.12 | — |
| AlgoTune | 1.7 | — |

## Agentic & Tool Use

- DeepSeek-R1: 30.7 (#75)
- Mistral Medium 3.5: —

| Benchmark | DeepSeek-R1 | Mistral Medium 3.5 |
|---|---|---|
| DeepResearch Bench | 35.1% | — |
| BALROG | 34.9% | — |
| METR Time Horizons | 53.8% | — |

## Reasoning

- DeepSeek-R1: 18.6 (#278)
- Mistral Medium 3.5: 17.3 (#295)

| Benchmark | DeepSeek-R1 | Mistral Medium 3.5 |
|---|---|---|
| Kagi LLM Benchmark | 69.4% | 41.4% |
| LMArena Hard Prompts | 1416 | 1436 |
| Epoch Capabilities Index | 141.29 | 141.35 |
| ARC-AGI-2 | 1.3% | — |
| SimpleBench | 40.8% | — |
| NYT Connections (extended) | — | 12.9% |
| ARC-AGI-1 | 21.2% | — |
| CritPt | 1.1% | — |
| LiveBench Reasoning | 83.2% | — |
| LiveBench Data Analysis | 69.8% | — |
| ForecastBench | 60 | — |
| LiveBench | 71.6% | — |

## Math

- DeepSeek-R1: 43.8 (#79)
- Mistral Medium 3.5: 39.1 (#113)

| Benchmark | DeepSeek-R1 | Mistral Medium 3.5 |
|---|---|---|
| LMArena Math | 1400 | 1431 |
| OTIS Mock AIME 2024-2025 | 66.4% | — |
| Omni-MATH | 42.4% | — |
| LiveBench Math | 80.7% | — |
| MATH Level 5 | 96.6% | — |

## Knowledge

- DeepSeek-R1: 44.5 (#87)
- Mistral Medium 3.5: 40.0 (#126)

| Benchmark | DeepSeek-R1 | Mistral Medium 3.5 |
|---|---|---|
| LMArena Expert | 1394 | 1432 |
| GPQA Diamond | 76.3% | — |
| MMLU-Pro | 79.3% | — |
| Confabulations | 12.7% | — |
| Vectara Hallucination Rate | 11.3% | — |
| GPQA (HELM) | 66.6% | — |

## Multimodal

- DeepSeek-R1: —
- Mistral Medium 3.5: 38.3 (#65)

| Benchmark | DeepSeek-R1 | Mistral Medium 3.5 |
|---|---|---|
| LMArena Vision | — | 1223 |

## Multilingual

- DeepSeek-R1: 52.4 (#85)
- Mistral Medium 3.5: 51.9 (#100)

| Benchmark | DeepSeek-R1 | Mistral Medium 3.5 |
|---|---|---|
| LMArena Non-English | 1412 | 1404 |
| LMArena Chinese | 1442 | 1442 |
| LMArena French | 1417 | 1448 |
| LMArena German | 1404 | 1451 |
| LMArena Korean | 1360 | 1385 |
| LMArena Russian | 1423 | 1395 |
| LMArena Spanish | 1411 | 1409 |
| LMArena Japanese | 1391 | — |

## Instruction Following

- DeepSeek-R1: 72.0 (#143)
- Mistral Medium 3.5: 74.6 (#90)

| Benchmark | DeepSeek-R1 | Mistral Medium 3.5 |
|---|---|---|
| LMArena Instruction Following | 1382 | 1415 |
| LiveBench Instruction Following | 80.5% | — |
| IFEval | 78.4% | — |

## Long Context

- DeepSeek-R1: 45.4 (#36)
- Mistral Medium 3.5: 43.2 (#103)

| Benchmark | DeepSeek-R1 | Mistral Medium 3.5 |
|---|---|---|
| LMArena Longer Query | 1391 | 1415 |
| Fiction.LiveBench | 75% | — |

## Writing & Preference

- DeepSeek-R1: 61.4 (#88)
- Mistral Medium 3.5: 58.5 (#117)

| Benchmark | DeepSeek-R1 | Mistral Medium 3.5 |
|---|---|---|
| LMArena Text | 1428 | 1421 |
| LMArena Creative Writing | 1405 | 1374 |
| LMArena Multi-Turn | 1405 | 1423 |
| Short-Story Creative Writing | 83% | — |
| EQ-Bench Creative Writing | 1500 | — |
| WildBench | 82.8% | — |
| EQ-Bench 4 | — | 993 |
| LiveBench Language | 48.5% | — |

## FAQ

### Is DeepSeek-R1 better than Mistral Medium 3.5?

DeepSeek-R1 is the stronger model overall, scoring 42.3 to 40.2 on the Noometry Index.

### Which is cheaper, DeepSeek-R1 or Mistral Medium 3.5?

DeepSeek-R1 is cheaper. It lists at $0.50 per million input tokens and $2.15 per million output tokens; Mistral Medium 3.5 lists at $1.50 and $7.50.

### Is DeepSeek-R1 or Mistral Medium 3.5 better for coding?

DeepSeek-R1 scores higher on coding benchmarks: 46.3 versus 36.0 in the Noometry coding category.

### Which has the bigger context window?

Mistral Medium 3.5 does, with 262K tokens against 164K.

### How many benchmarks do DeepSeek-R1 and Mistral Medium 3.5 share?

18 benchmarks have published results for both models. DeepSeek-R1 has 52 scored results on Noometry and Mistral Medium 3.5 has 22.