Model comparison
Deepseek Coder v2 vs Mistral Small 3.1
Deepseek Coder v2 is the stronger model overall, scoring 35.9 to 31.7 on the Noometry Index.
Last verified . 17 shared benchmarks.
DeepSeek
Mistral AISummary
- They share 17 benchmarks with published results for both. Deepseek Coder v2 scores higher in 4 categories and Mistral Small 3.1 in 4 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Deepseek Coder v2 leads 34.9 to 14.7.
Side by side
| Deepseek Coder v2 | Mistral Small 3.1 | |
|---|---|---|
| Provider | DeepSeek | Mistral AI |
| Noometry Index | 35.9 | 31.7 |
| Released | 2024-06-17 | 2025-03-17 |
| Weights | Open | Open |
| Context window | — | 128K |
| Max output | — | 102K |
| Input $ / M tokens | — | $0.35 |
| Output $ / M tokens | — | $0.56 |
| Results tracked | 24 | 28 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Deepseek Coder v2: 38.1 (#183), Mistral Small 3.1: 38.3 (#179)
| Benchmark | Deepseek Coder v2 | Mistral Small 3.1 |
|---|---|---|
| LMArena Coding | 1251 | 1309 |
| BigCodeBench Instruct | 48.2% | — |
| BigCodeBench Complete | 59.7% | — |
| HumanEval+ | 82.3% | — |
| MBPP+ | 75.1% | — |
Reasoning Deepseek Coder v2 leads
Deepseek Coder v2: 23.6 (#176), Mistral Small 3.1: 19.7 (#254)
| Benchmark | Deepseek Coder v2 | Mistral Small 3.1 |
|---|---|---|
| LMArena Hard Prompts | 1207 | 1278 |
| Chess Puzzles | — | 1% |
| Epoch Capabilities Index | — | 127.48 |
| WinoGrande | 83.7% | — |
Math Deepseek Coder v2 leads
Deepseek Coder v2: 34.9 (#190), Mistral Small 3.1: 14.7 (#301)
| Benchmark | Deepseek Coder v2 | Mistral Small 3.1 |
|---|---|---|
| LMArena Math | 1241 | 1262 |
| OTIS Mock AIME 2024-2025 | — | 3.9% |
| Omni-MATH | — | 24.8% |
| GSM8K | 94.5% | — |
Knowledge Deepseek Coder v2 leads
Deepseek Coder v2: 32.3 (#212), Mistral Small 3.1: 22.6 (#271)
| Benchmark | Deepseek Coder v2 | Mistral Small 3.1 |
|---|---|---|
| LMArena Expert | 1181 | 1257 |
| GPQA Diamond | — | 41.9% |
| MMLU-Pro | — | 61% |
| GPQA (HELM) | — | 39.2% |
| ARC (AI2) Challenge | 64.3% | — |
Multimodal Not comparable
Deepseek Coder v2: —, Mistral Small 3.1: 33.2 (#99)
| Benchmark | Deepseek Coder v2 | Mistral Small 3.1 |
|---|---|---|
| LMArena Vision | — | 1136 |
Multilingual Mistral Small 3.1 leads
Deepseek Coder v2: 36.3 (#240), Mistral Small 3.1: 41.2 (#209)
| Benchmark | Deepseek Coder v2 | Mistral Small 3.1 |
|---|---|---|
| LMArena Non-English | 1182 | 1255 |
| LMArena Chinese | 1201 | 1253 |
| LMArena French | 1185 | 1273 |
| LMArena German | 1164 | 1266 |
| LMArena Japanese | 1126 | 1208 |
| LMArena Korean | 1104 | 1206 |
| LMArena Russian | 1188 | 1263 |
| LMArena Spanish | 1153 | 1283 |
Instruction Following Mistral Small 3.1 leads
Deepseek Coder v2: 61.7 (#242), Mistral Small 3.1: 63.6 (#230)
| Benchmark | Deepseek Coder v2 | Mistral Small 3.1 |
|---|---|---|
| LMArena Instruction Following | 1180 | 1264 |
| IFEval | — | 75% |
Long Context Mistral Small 3.1 leads
Deepseek Coder v2: 37.0 (#224), Mistral Small 3.1: 39.5 (#178)
| Benchmark | Deepseek Coder v2 | Mistral Small 3.1 |
|---|---|---|
| LMArena Longer Query | 1219 | 1299 |
Writing & Preference Deepseek Coder v2 leads
Deepseek Coder v2: 38.2 (#253), Mistral Small 3.1: 37.0 (#259)
| Benchmark | Deepseek Coder v2 | Mistral Small 3.1 |
|---|---|---|
| LMArena Text | 1191 | 1277 |
| LMArena Creative Writing | 1120 | 1253 |
| LMArena Multi-Turn | 1177 | 1270 |
| EQ-Bench Creative Writing | — | 761 |
| WildBench | — | 78.8% |
Frequently asked questions
Is Deepseek Coder v2 better than Mistral Small 3.1?
Deepseek Coder v2 is the stronger model overall, scoring 35.9 to 31.7 on the Noometry Index.
Is Deepseek Coder v2 or Mistral Small 3.1 better for coding?
They score almost the same on coding (38.1 vs 38.3); test both on your own repository before choosing.
How many benchmarks do Deepseek Coder v2 and Mistral Small 3.1 share?
17 benchmarks have published results for both models. Deepseek Coder v2 has 24 scored results on Noometry and Mistral Small 3.1 has 28.