Model comparison
Deepseek Coder v2 vs Wizardlm 70b
Deepseek Coder v2 is the stronger model overall, scoring 35.9 to 33.0 on the Noometry Index.
Last verified . 12 shared benchmarks.
DeepSeek
MicrosoftSummary
- They share 12 benchmarks with published results for both. Deepseek Coder v2 scores higher in 7 categories and Wizardlm 70b in 0 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in coding, where Deepseek Coder v2 leads 38.1 to 31.4.
Side by side
| Deepseek Coder v2 | Wizardlm 70b | |
|---|---|---|
| Provider | DeepSeek | Microsoft |
| Noometry Index | 35.9 | 33.0 |
| Released | 2024-06-17 | — |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 24 | 12 |
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), Wizardlm 70b: 31.4 (#285)
| Benchmark | Deepseek Coder v2 | Wizardlm 70b |
|---|---|---|
| LMArena Coding | 1251 | 1081 |
| 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), Wizardlm 70b: 20.7 (#234)
| Benchmark | Deepseek Coder v2 | Wizardlm 70b |
|---|---|---|
| LMArena Hard Prompts | 1207 | 1079 |
| WinoGrande | 83.7% | — |
Math Deepseek Coder v2 leads
Deepseek Coder v2: 34.9 (#190), Wizardlm 70b: 32.2 (#218)
| Benchmark | Deepseek Coder v2 | Wizardlm 70b |
|---|---|---|
| LMArena Math | 1241 | 1116 |
| GSM8K | 94.5% | — |
Knowledge Not comparable
Deepseek Coder v2: 32.3 (#212), Wizardlm 70b: —
| Benchmark | Deepseek Coder v2 | Wizardlm 70b |
|---|---|---|
| LMArena Expert | 1181 | — |
| ARC (AI2) Challenge | 64.3% | — |
Multilingual Deepseek Coder v2 leads
Deepseek Coder v2: 36.3 (#240), Wizardlm 70b: 29.6 (#265)
| Benchmark | Deepseek Coder v2 | Wizardlm 70b |
|---|---|---|
| LMArena Non-English | 1182 | 1078 |
| LMArena Chinese | 1201 | 1052 |
| LMArena German | 1164 | 1083 |
| LMArena Russian | 1188 | 1155 |
| LMArena French | 1185 | — |
| LMArena Japanese | 1126 | — |
| LMArena Korean | 1104 | — |
| LMArena Spanish | 1153 | — |
Instruction Following Deepseek Coder v2 leads
Deepseek Coder v2: 61.7 (#242), Wizardlm 70b: 56.3 (#273)
| Benchmark | Deepseek Coder v2 | Wizardlm 70b |
|---|---|---|
| LMArena Instruction Following | 1180 | 1093 |
Long Context Deepseek Coder v2 leads
Deepseek Coder v2: 37.0 (#224), Wizardlm 70b: 33.3 (#263)
| Benchmark | Deepseek Coder v2 | Wizardlm 70b |
|---|---|---|
| LMArena Longer Query | 1219 | 1097 |
Writing & Preference Deepseek Coder v2 leads
Deepseek Coder v2: 38.2 (#253), Wizardlm 70b: 34.8 (#269)
| Benchmark | Deepseek Coder v2 | Wizardlm 70b |
|---|---|---|
| LMArena Text | 1191 | 1120 |
| LMArena Creative Writing | 1120 | 1149 |
| LMArena Multi-Turn | 1177 | 1108 |
Frequently asked questions
Is Deepseek Coder v2 better than Wizardlm 70b?
Deepseek Coder v2 is the stronger model overall, scoring 35.9 to 33.0 on the Noometry Index.
Is Deepseek Coder v2 or Wizardlm 70b better for coding?
Deepseek Coder v2 scores higher on coding benchmarks: 38.1 versus 31.4 in the Noometry coding category.
How many benchmarks do Deepseek Coder v2 and Wizardlm 70b share?
12 benchmarks have published results for both models. Deepseek Coder v2 has 24 scored results on Noometry and Wizardlm 70b has 12.