Model comparison
DeepSeek-R1 vs Mistral Medium
DeepSeek-R1 is the stronger model overall, scoring 42.3 to 36.3 on the Noometry Index.
Last verified . 27 shared benchmarks.
DeepSeek
Mistral AISummary
- They share 27 benchmarks with published results for both. DeepSeek-R1 scores higher in 7 categories and Mistral Medium in 2 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where DeepSeek-R1 leads 44.5 to 25.0.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 66.4% for DeepSeek-R1 and 32.2% for Mistral Medium.
- DeepSeek-R1 is cheaper at $0.50 / $2.15 per million input/output tokens, against $1.50 / $7.50 for Mistral Medium.
- Mistral Medium accepts more context: 262K tokens versus 164K.
- Mistral Medium has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-R1 | Mistral Medium | |
|---|---|---|
| Provider | DeepSeek | Mistral AI |
| Noometry Index | 42.3 | 36.3 |
| Released | 2025-01-20 | 2023-12-11 |
| Weights | Proprietary | Open |
| Context window | 164K | 262K |
| Max output | 64K | 262K |
| Input $ / M tokens | $0.50 | $1.50 |
| Output $ / M tokens | $2.15 | $7.50 |
| Results tracked | 52 | 36 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-R1 leads
DeepSeek-R1: 46.3 (#68), Mistral Medium: 34.2 (#243)
| Benchmark | DeepSeek-R1 | Mistral Medium |
|---|---|---|
| SciCode | 35.7% | 40.2% |
| WeirdML | 41.6% | 43.7% |
| LMArena Coding | 1427 | 1434 |
| ALE-Bench | 804.12 | 763.98 |
| FrontierCode | — | 8% |
| Aider Polyglot | 71.4% | — |
| LiveBench Coding | 66.7% | — |
| AlgoTune | 1.7 | — |
Agentic & Tool Use DeepSeek-R1 leads
DeepSeek-R1: 30.7 (#75), Mistral Medium: 28.3 (#90)
| Benchmark | DeepSeek-R1 | Mistral Medium |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 37.7% |
| DeepResearch Bench | 35.1% | — |
| BALROG | 34.9% | — |
| METR Time Horizons | 53.8% | — |
Reasoning Mistral Medium leads
DeepSeek-R1: 18.6 (#278), Mistral Medium: 24.0 (#167)
| Benchmark | DeepSeek-R1 | Mistral Medium |
|---|---|---|
| Kagi LLM Benchmark | 69.4% | 50% |
| CritPt | 1.1% | 0% |
| LMArena Hard Prompts | 1416 | 1426 |
| ARC-AGI-2 | 1.3% | — |
| SimpleBench | 40.8% | — |
| ARC-AGI-1 | 21.2% | — |
| LiveBench Reasoning | 83.2% | — |
| DTBench | — | 75.5% |
| LiveBench Data Analysis | 69.8% | — |
| LMCA | — | 26.1% |
| Surface Evolver Bench | — | 26.9% |
| Epoch Capabilities Index | 141.29 | — |
| ForecastBench | 60 | — |
| LiveBench | 71.6% | — |
Math DeepSeek-R1 leads
DeepSeek-R1: 43.8 (#79), Mistral Medium: 28.1 (#245)
| Benchmark | DeepSeek-R1 | Mistral Medium |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 66.4% | 32.2% |
| LMArena Math | 1400 | 1408 |
| MATH Level 5 | 96.6% | 81.6% |
| ProofBench | — | 9% |
| Omni-MATH | 42.4% | — |
| LiveBench Math | 80.7% | — |
| FrontierMath (Feb 2025 set) | — | 0.3% |
Knowledge DeepSeek-R1 leads
DeepSeek-R1: 44.5 (#87), Mistral Medium: 25.0 (#265)
| Benchmark | DeepSeek-R1 | Mistral Medium |
|---|---|---|
| GPQA Diamond | 76.3% | 59.5% |
| Vectara Hallucination Rate | 11.3% | 22.7% |
| LMArena Expert | 1394 | 1408 |
| Humanity's Last Exam | — | 4.5% |
| MMLU-Pro | 79.3% | — |
| Confabulations | 12.7% | — |
| GPQA (HELM) | 66.6% | — |
Multimodal Not comparable
DeepSeek-R1: —, Mistral Medium: 35.3 (#88)
| Benchmark | DeepSeek-R1 | Mistral Medium |
|---|---|---|
| LMArena Vision | — | 1172 |
Multilingual Too close to call
DeepSeek-R1: 52.4 (#85), Mistral Medium: 52.1 (#91)
| Benchmark | DeepSeek-R1 | Mistral Medium |
|---|---|---|
| LMArena Non-English | 1412 | 1408 |
| LMArena Chinese | 1442 | 1447 |
| LMArena French | 1417 | 1459 |
| LMArena German | 1404 | 1432 |
| LMArena Japanese | 1391 | 1378 |
| LMArena Korean | 1360 | 1380 |
| LMArena Russian | 1423 | 1411 |
| LMArena Spanish | 1411 | 1433 |
Instruction Following Mistral Medium leads
DeepSeek-R1: 72.0 (#143), Mistral Medium: 73.7 (#116)
| Benchmark | DeepSeek-R1 | Mistral Medium |
|---|---|---|
| LMArena Instruction Following | 1382 | 1398 |
| LiveBench Instruction Following | 80.5% | — |
| IFEval | 78.4% | — |
Long Context DeepSeek-R1 leads
DeepSeek-R1: 45.4 (#36), Mistral Medium: 42.9 (#114)
| Benchmark | DeepSeek-R1 | Mistral Medium |
|---|---|---|
| LMArena Longer Query | 1391 | 1406 |
| Fiction.LiveBench | 75% | — |
Writing & Preference DeepSeek-R1 leads
DeepSeek-R1: 61.4 (#88), Mistral Medium: 60.0 (#103)
| Benchmark | DeepSeek-R1 | Mistral Medium |
|---|---|---|
| LMArena Text | 1428 | 1424 |
| LMArena Creative Writing | 1405 | 1391 |
| Short-Story Creative Writing | 83% | 77.3% |
| LMArena Multi-Turn | 1405 | 1418 |
| EQ-Bench Creative Writing | 1500 | — |
| WildBench | 82.8% | — |
| LiveBench Language | 48.5% | — |
Frequently asked questions
Is DeepSeek-R1 better than Mistral Medium?
DeepSeek-R1 is the stronger model overall, scoring 42.3 to 36.3 on the Noometry Index.
Which is cheaper, DeepSeek-R1 or Mistral Medium?
DeepSeek-R1 is cheaper. It lists at $0.50 per million input tokens and $2.15 per million output tokens; Mistral Medium lists at $1.50 and $7.50.
Is DeepSeek-R1 or Mistral Medium better for coding?
DeepSeek-R1 scores higher on coding benchmarks: 46.3 versus 34.2 in the Noometry coding category.
Which has the bigger context window?
Mistral Medium does, with 262K tokens against 164K.
How many benchmarks do DeepSeek-R1 and Mistral Medium share?
27 benchmarks have published results for both models. DeepSeek-R1 has 52 scored results on Noometry and Mistral Medium has 36.