Model comparison
Deepseek Coder v2 vs Llama 2-70B
Deepseek Coder v2 is the stronger model overall, scoring 35.9 to 24.4 on the Noometry Index.
Last verified . 20 shared benchmarks.
DeepSeek
MetaSummary
- They share 20 benchmarks with published results for both. Deepseek Coder v2 scores higher in 8 categories and Llama 2-70B 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.
Side by side
| Deepseek Coder v2 | Llama 2-70B | |
|---|---|---|
| Provider | DeepSeek | Meta |
| Noometry Index | 35.9 | 24.4 |
| Released | 2024-06-17 | 2023-07-18 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 24 | 35 |
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), Llama 2-70B: 31.4 (#286)
| Benchmark | Deepseek Coder v2 | Llama 2-70B |
|---|---|---|
| LMArena Coding | 1251 | 1079 |
| 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), Llama 2-70B: 14.4 (#325)
| Benchmark | Deepseek Coder v2 | Llama 2-70B |
|---|---|---|
| LMArena Hard Prompts | 1207 | 1073 |
| WinoGrande | 83.7% | 80.2% |
| DTBench | — | 41.6% |
| BIG-Bench Hard | — | 64.9% |
| CommonsenseQA 2.0 | — | 50% |
| Epoch Capabilities Index | — | 113.79 |
| ForecastBench | — | 51.4 |
| HellaSwag | — | 85.3% |
| LAMBADA | — | 78.9% |
| PIQA | — | 82.8% |
Math Deepseek Coder v2 leads
Deepseek Coder v2: 34.9 (#190), Llama 2-70B: 8.1 (#326)
| Benchmark | Deepseek Coder v2 | Llama 2-70B |
|---|---|---|
| LMArena Math | 1241 | 1091 |
| GSM8K | 94.5% | 69.6% |
| OTIS Mock AIME 2024-2025 | — | 0% |
| MATH Level 5 | — | 3.3% |
Knowledge Deepseek Coder v2 leads
Deepseek Coder v2: 32.3 (#212), Llama 2-70B: 7.4 (#310)
| Benchmark | Deepseek Coder v2 | Llama 2-70B |
|---|---|---|
| LMArena Expert | 1181 | 1039 |
| ARC (AI2) Challenge | 64.3% | 78.3% |
| GPQA Diamond | — | 26.3% |
| BoolQ | — | 88.6% |
| MMLU | — | 69.9% |
| OpenBookQA | — | 60.2% |
| TriviaQA | — | 87.6% |
Multilingual Deepseek Coder v2 leads
Deepseek Coder v2: 36.3 (#240), Llama 2-70B: 27.7 (#274)
| Benchmark | Deepseek Coder v2 | Llama 2-70B |
|---|---|---|
| LMArena Non-English | 1182 | 1045 |
| LMArena Chinese | 1201 | 995 |
| LMArena French | 1185 | 1090 |
| LMArena German | 1164 | 1041 |
| LMArena Japanese | 1126 | 927 |
| LMArena Korean | 1104 | 964 |
| LMArena Russian | 1188 | 1083 |
| LMArena Spanish | 1153 | 1143 |
Instruction Following Deepseek Coder v2 leads
Deepseek Coder v2: 61.7 (#242), Llama 2-70B: 54.9 (#278)
| Benchmark | Deepseek Coder v2 | Llama 2-70B |
|---|---|---|
| LMArena Instruction Following | 1180 | 1071 |
Long Context Deepseek Coder v2 leads
Deepseek Coder v2: 37.0 (#224), Llama 2-70B: 32.3 (#270)
| Benchmark | Deepseek Coder v2 | Llama 2-70B |
|---|---|---|
| LMArena Longer Query | 1219 | 1062 |
Writing & Preference Deepseek Coder v2 leads
Deepseek Coder v2: 38.2 (#253), Llama 2-70B: 32.3 (#279)
| Benchmark | Deepseek Coder v2 | Llama 2-70B |
|---|---|---|
| LMArena Text | 1191 | 1115 |
| LMArena Creative Writing | 1120 | 1075 |
| LMArena Multi-Turn | 1177 | 1088 |
Frequently asked questions
Is Deepseek Coder v2 better than Llama 2-70B?
Deepseek Coder v2 is the stronger model overall, scoring 35.9 to 24.4 on the Noometry Index.
Is Deepseek Coder v2 or Llama 2-70B 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 Llama 2-70B share?
20 benchmarks have published results for both models. Deepseek Coder v2 has 24 scored results on Noometry and Llama 2-70B has 35.