DeepSeek Coder 6.7B
DeepSeek Coder 6.7B.
5
Benchmarks with a score
0
Ranked first, out of five or more models
0
In the top three
DeepSeek Nov 2023 DeepSeek Coder 6.7B
Open weights
67.8
#250 of 256
Epoch Capabilities Index
- AutoCodeBench 20.5% #32 /34
Where it ranks
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
Knowledge and language
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MMLU | 36.4% | #73 of 85 | Claim in Qwen2.5-Coder Technical Report, via Epoch AI | |
| ARC-Challenge | 36.4% | #65 of 75 | Claim in Qwen2.5-Coder Technical Report, via Epoch AI | |
| WinoGrande | 57.6% | #70 of 76 | Claim in Qwen2.5-Coder Technical Report, via Epoch AI |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| GSM8K | 21.3% | #59 of 70 | Claim in Qwen2.5-Coder Technical Report, via Epoch AI |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| AutoCodeBench | 20.5% | #32 of 34 | Official leaderboard |
DeepSeek's models over time
DeepSeek's models in color, against every other model on the Epoch Capabilities Index, by release date.
Source: Epoch Capabilities Index, Epoch AI (CC BY 4.0)