Model comparison
Deepseek Coder v2 vs Mistral 7B
Deepseek Coder v2 is the stronger model overall, scoring 35.9 to 23.0 on the Noometry Index.
Last verified . 23 shared benchmarks.
DeepSeek
Mistral AISummary
- They share 23 benchmarks with published results for both. Deepseek Coder v2 scores higher in 8 categories and Mistral 7B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Deepseek Coder v2 leads 34.9 to 8.1.
- The biggest single-benchmark swing is BigCodeBench Complete: 59.7% for Deepseek Coder v2 and 27.3% for Mistral 7B.
Side by side
| Deepseek Coder v2 | Mistral 7B | |
|---|---|---|
| Provider | DeepSeek | Mistral AI |
| Noometry Index | 35.9 | 23.0 |
| Released | 2024-06-17 | 2023-09-27 |
| Weights | Open | Open |
| Context window | — | 8K |
| Max output | — | 8K |
| Input $ / M tokens | — | $0.25 |
| Output $ / M tokens | — | $0.25 |
| Results tracked | 24 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Deepseek Coder v2 leads
Deepseek Coder v2: 38.1 (#183), Mistral 7B: 26.4 (#326)
| Benchmark | Deepseek Coder v2 | Mistral 7B |
|---|---|---|
| BigCodeBench Instruct | 48.2% | 19.5% |
| LMArena Coding | 1251 | 1082 |
| BigCodeBench Complete | 59.7% | 27.3% |
| HumanEval+ | 82.3% | 36% |
| MBPP+ | 75.1% | 42.1% |
Reasoning Deepseek Coder v2 leads
Deepseek Coder v2: 23.6 (#176), Mistral 7B: 13.1 (#336)
| Benchmark | Deepseek Coder v2 | Mistral 7B |
|---|---|---|
| LMArena Hard Prompts | 1207 | 1067 |
| WinoGrande | 83.7% | 75.3% |
| Chess Puzzles | — | 0% |
| DTBench | — | 42.5% |
| Adversarial NLI | — | 47.1% |
| BIG-Bench Hard | — | 56.1% |
| Epoch Capabilities Index | — | 112.21 |
| HellaSwag | — | 81% |
| PIQA | — | 83% |
Math Deepseek Coder v2 leads
Deepseek Coder v2: 34.9 (#190), Mistral 7B: 8.1 (#325)
| Benchmark | Deepseek Coder v2 | Mistral 7B |
|---|---|---|
| LMArena Math | 1241 | 1085 |
| GSM8K | 94.5% | 54.4% |
| OTIS Mock AIME 2024-2025 | — | 0.3% |
| MATH Level 5 | — | 3.7% |
Knowledge Deepseek Coder v2 leads
Deepseek Coder v2: 32.3 (#212), Mistral 7B: 7.4 (#311)
| Benchmark | Deepseek Coder v2 | Mistral 7B |
|---|---|---|
| LMArena Expert | 1181 | 1036 |
| ARC (AI2) Challenge | 64.3% | 78.6% |
| GPQA Diamond | — | 15.2% |
| BoolQ | — | 87.4% |
| MMLU | — | 62.5% |
| OpenBookQA | — | 79.8% |
| TriviaQA | — | 75.2% |
Multilingual Deepseek Coder v2 leads
Deepseek Coder v2: 36.3 (#240), Mistral 7B: 25.8 (#283)
| Benchmark | Deepseek Coder v2 | Mistral 7B |
|---|---|---|
| LMArena Non-English | 1182 | 1012 |
| LMArena Chinese | 1201 | 1009 |
| LMArena French | 1185 | 1037 |
| LMArena German | 1164 | 987 |
| LMArena Japanese | 1126 | 878 |
| LMArena Russian | 1188 | 1018 |
| LMArena Spanish | 1153 | 1026 |
| LMArena Korean | 1104 | — |
Instruction Following Deepseek Coder v2 leads
Deepseek Coder v2: 61.7 (#242), Mistral 7B: 54.2 (#280)
| Benchmark | Deepseek Coder v2 | Mistral 7B |
|---|---|---|
| LMArena Instruction Following | 1180 | 1060 |
Long Context Deepseek Coder v2 leads
Deepseek Coder v2: 37.0 (#224), Mistral 7B: 32.2 (#271)
| Benchmark | Deepseek Coder v2 | Mistral 7B |
|---|---|---|
| LMArena Longer Query | 1219 | 1060 |
Writing & Preference Deepseek Coder v2 leads
Deepseek Coder v2: 38.2 (#253), Mistral 7B: 30.7 (#286)
| Benchmark | Deepseek Coder v2 | Mistral 7B |
|---|---|---|
| LMArena Text | 1191 | 1090 |
| LMArena Creative Writing | 1120 | 1068 |
| LMArena Multi-Turn | 1177 | 1062 |
Frequently asked questions
Is Deepseek Coder v2 better than Mistral 7B?
Deepseek Coder v2 is the stronger model overall, scoring 35.9 to 23.0 on the Noometry Index.
Is Deepseek Coder v2 or Mistral 7B better for coding?
Deepseek Coder v2 scores higher on coding benchmarks: 38.1 versus 26.4 in the Noometry coding category.
How many benchmarks do Deepseek Coder v2 and Mistral 7B share?
23 benchmarks have published results for both models. Deepseek Coder v2 has 24 scored results on Noometry and Mistral 7B has 37.