Model comparison
DeepSeek Coder 6.7B vs Llama 3-70B
Llama 3-70B has enough public results to be ranked (#323); DeepSeek Coder 6.7B does not yet, so treat this comparison as directional.
Last verified . 7 shared benchmarks.
DeepSeek
MetaSummary
- They share 7 benchmarks with published results for both. DeepSeek Coder 6.7B scores higher in 0 categories and Llama 3-70B in 1 category, but none of those gaps is larger than the uncertainty.
- The biggest single-benchmark swing is BigCodeBench Complete: 43.8% for DeepSeek Coder 6.7B and 54.5% for Llama 3-70B.
Side by side
| DeepSeek Coder 6.7B | Llama 3-70B | |
|---|---|---|
| Provider | DeepSeek | Meta |
| Noometry Index | 37.6 | 28.8 |
| Released | 2023-11-02 | 2024-04-18 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 9 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
DeepSeek Coder 6.7B: 35.8 (#219), Llama 3-70B: 35.8 (#218)
| Benchmark | DeepSeek Coder 6.7B | Llama 3-70B |
|---|---|---|
| BigCodeBench Instruct | 35.5% | 43.6% |
| BigCodeBench Complete | 43.8% | 54.5% |
| HumanEval+ | 71.3% | 72% |
| MBPP+ | 65.6% | 69% |
| LMArena Coding | — | 1206 |
Agentic & Tool Use Not comparable
DeepSeek Coder 6.7B: —, Llama 3-70B: 21.1 (#139)
| Benchmark | DeepSeek Coder 6.7B | Llama 3-70B |
|---|---|---|
| Cybench | — | 5% |
Reasoning Not comparable
DeepSeek Coder 6.7B: —, Llama 3-70B: 18.0 (#288)
| Benchmark | DeepSeek Coder 6.7B | Llama 3-70B |
|---|---|---|
| Epoch Capabilities Index | 89.62 | 122.93 |
| WinoGrande | 57.6% | 83.5% |
| Kagi LLM Benchmark | — | 35.1% |
| LMArena Hard Prompts | — | 1195 |
| DTBench | — | 54.2% |
| ForecastBench | — | 57.1 |
Math Not comparable
DeepSeek Coder 6.7B: —, Llama 3-70B: 12.8 (#305)
| Benchmark | DeepSeek Coder 6.7B | Llama 3-70B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 4.3% |
| LMArena Math | — | 1218 |
| MATH Level 5 | — | 22.6% |
| GSM8K | 21.3% | — |
Knowledge Not comparable
DeepSeek Coder 6.7B: —, Llama 3-70B: 20.8 (#277)
| Benchmark | DeepSeek Coder 6.7B | Llama 3-70B |
|---|---|---|
| MMLU | 36.4% | 79.3% |
| GPQA Diamond | — | 40.6% |
| LMArena Expert | — | 1149 |
| ARC (AI2) Challenge | 36.4% | — |
Multilingual Not comparable
DeepSeek Coder 6.7B: —, Llama 3-70B: 33.6 (#251)
| Benchmark | DeepSeek Coder 6.7B | Llama 3-70B |
|---|---|---|
| LMArena Non-English | — | 1142 |
| LMArena Chinese | — | 1114 |
| LMArena French | — | 1232 |
| LMArena German | — | 1169 |
| LMArena Japanese | — | 1017 |
| LMArena Korean | — | 1017 |
| LMArena Russian | — | 1159 |
| LMArena Spanish | — | 1241 |
Instruction Following Not comparable
DeepSeek Coder 6.7B: —, Llama 3-70B: 62.5 (#238)
| Benchmark | DeepSeek Coder 6.7B | Llama 3-70B |
|---|---|---|
| LMArena Instruction Following | — | 1194 |
Long Context Not comparable
DeepSeek Coder 6.7B: —, Llama 3-70B: 35.6 (#240)
| Benchmark | DeepSeek Coder 6.7B | Llama 3-70B |
|---|---|---|
| LMArena Longer Query | — | 1174 |
Writing & Preference Not comparable
DeepSeek Coder 6.7B: —, Llama 3-70B: 42.8 (#231)
| Benchmark | DeepSeek Coder 6.7B | Llama 3-70B |
|---|---|---|
| LMArena Text | — | 1221 |
| LMArena Creative Writing | — | 1210 |
| LMArena Multi-Turn | — | 1223 |
Frequently asked questions
Is DeepSeek Coder 6.7B better than Llama 3-70B?
Llama 3-70B has enough public results to be ranked (#323); DeepSeek Coder 6.7B does not yet, so treat this comparison as directional.
Is DeepSeek Coder 6.7B or Llama 3-70B better for coding?
They score almost the same on coding (35.8 vs 35.8); test both on your own repository before choosing.
How many benchmarks do DeepSeek Coder 6.7B and Llama 3-70B share?
7 benchmarks have published results for both models. DeepSeek Coder 6.7B has 9 scored results on Noometry and Llama 3-70B has 31.