Model comparison
DeepSeek-V2.5 (Sep 2024) vs Magistral Medium
DeepSeek-V2.5 (Sep 2024) is the stronger model overall, scoring 37.6 to 35.2 on the Noometry Index.
Last verified . 17 shared benchmarks.
DeepSeek
Mistral AISummary
- They share 17 benchmarks with published results for both. DeepSeek-V2.5 (Sep 2024) scores higher in 7 categories and Magistral Medium in 1 category; 6 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek-V2.5 (Sep 2024) leads 25.6 to 8.6.
Side by side
| DeepSeek-V2.5 (Sep 2024) | Magistral Medium | |
|---|---|---|
| Provider | DeepSeek | Mistral AI |
| Noometry Index | 37.6 | 35.2 |
| Released | 2024-09-06 | 2025-03-17 |
| Weights | Open | Open |
| Context window | — | 262K |
| Max output | — | 16K |
| Input $ / M tokens | — | $2 |
| Output $ / M tokens | — | $5 |
| Results tracked | 22 | 22 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Magistral Medium leads
DeepSeek-V2.5 (Sep 2024): 31.7 (#281), Magistral Medium: 39.1 (#161)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Magistral Medium |
|---|---|---|
| LMArena Coding | 1309 | 1319 |
| Aider Polyglot | 17.8% | — |
| SciCode | — | 39.2% |
| BigCodeBench Instruct | 48.6% | — |
| BigCodeBench Complete | 53.2% | — |
| HumanEval+ | 83.5% | — |
| MBPP+ | 74.1% | — |
Reasoning DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 25.6 (#145), Magistral Medium: 8.6 (#348)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Magistral Medium |
|---|---|---|
| LMArena Hard Prompts | 1289 | 1267 |
| ARC-AGI-2 | — | 0% |
| Kagi LLM Benchmark | — | 16.2% |
| ARC-AGI-1 | — | 6.1% |
| CritPt | — | 0.3% |
Math Too close to call
DeepSeek-V2.5 (Sep 2024): 35.9 (#177), Magistral Medium: 35.1 (#189)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Magistral Medium |
|---|---|---|
| LMArena Math | 1288 | 1250 |
Knowledge DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 34.8 (#193), Magistral Medium: 33.5 (#202)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Magistral Medium |
|---|---|---|
| LMArena Expert | 1266 | 1223 |
Multilingual DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 42.5 (#193), Magistral Medium: 39.6 (#224)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Magistral Medium |
|---|---|---|
| LMArena Non-English | 1273 | 1232 |
| LMArena Chinese | 1318 | 1227 |
| LMArena French | 1289 | 1267 |
| LMArena German | 1258 | 1248 |
| LMArena Japanese | 1228 | 1175 |
| LMArena Korean | 1209 | 1125 |
| LMArena Russian | 1289 | 1224 |
| LMArena Spanish | 1248 | 1271 |
Instruction Following DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 67.5 (#194), Magistral Medium: 66.0 (#211)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Magistral Medium |
|---|---|---|
| LMArena Instruction Following | 1280 | 1254 |
Long Context Too close to call
DeepSeek-V2.5 (Sep 2024): 39.5 (#174), Magistral Medium: 39.3 (#183)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Magistral Medium |
|---|---|---|
| LMArena Longer Query | 1301 | 1295 |
Writing & Preference DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 49.8 (#187), Magistral Medium: 46.3 (#219)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Magistral Medium |
|---|---|---|
| LMArena Text | 1294 | 1255 |
| LMArena Creative Writing | 1285 | 1245 |
| LMArena Multi-Turn | 1297 | 1275 |
Frequently asked questions
Is DeepSeek-V2.5 (Sep 2024) better than Magistral Medium?
DeepSeek-V2.5 (Sep 2024) is the stronger model overall, scoring 37.6 to 35.2 on the Noometry Index.
Is DeepSeek-V2.5 (Sep 2024) or Magistral Medium better for coding?
Magistral Medium scores higher on coding benchmarks: 39.1 versus 31.7 in the Noometry coding category.
How many benchmarks do DeepSeek-V2.5 (Sep 2024) and Magistral Medium share?
17 benchmarks have published results for both models. DeepSeek-V2.5 (Sep 2024) has 22 scored results on Noometry and Magistral Medium has 22.