DeepSeek·released 2024-06-131 source
DeepSeek-Coder-V2-Lite-Base benchmark scores: 4 benchmarks tracked. 0% of its scores are independently reproduced — the rest are self-reported or unverified.
Leads on 0 of 4 benchmarks·best result 67.1 on GSM8K (self-reported)·0 of 4 independently reproduced
| Benchmark | Score | Evidence status | Measured |
|---|---|---|---|
| GSM8K | 67.1accuracy (%) | self-reported· optimizedT1 | 2024-06-13Qwen2.5-Coder Technical Report |
| MMLU | 47.33accuracy (%) | self-reported· optimizedT1 | 2024-06-13Qwen2.5-Coder Technical Report |
| Winogrande | 45.8accuracy (%) | self-reported· optimizedT1 | 2024-06-13Qwen2.5-Coder Technical Report |
| ARC AI2 | 43.07accuracy (%) | self-reported· optimizedT1 | 2024-06-13Qwen2.5-Coder Technical Report |
Claims drawn from cited facts, not live model generation.
This model was developed by DeepSeek and released in 2024.
2 cited facts
At 4 benchmarks the coverage is a narrow slice; read the count as partially measured rather than as a full profile. Not topping any of them leaves it in the field, and since a top score is a single-harness artifact anyway, its absence is weak evidence about the model overall. Because the scores are self-reported, they should be treated as unverified claims rather than independently confirmed results.
3 cited facts
This model trails the leader by an average of 36.24 points, a wide gap that places it well behind the front of the pack. It holds the top score on none of the benchmarks, confirming it is not a leader in any category.
2 cited facts
1 facts cross-checked across data sources: 1 single-source.
Source facts, citations, and refresh stamp for this record.
Sources: epoch_benchmarks
DeepSeek-Coder-V2-Lite-Base is an AI model developed by DeepSeek, released 2024-06-13. Its benchmark record below tags every score as independently reproduced, self-reported, or unverified.
DeepSeek-Coder-V2-Lite-Base has recorded scores on 4 benchmarks, each shown with its evidence status.
DeepSeek-Coder-V2-Lite-Base has recorded scores on 4 benchmarks — Winogrande, MMLU, ARC AI2, GSM8K. The full table above shows each score with its evidence status.
0 of 4 recorded scores (0%) are independently reproduced rather than self-reported by the lab.
DeepSeek-Coder-V2-Lite-Base has 4 tracked claims: 4 self-reported.