Skip to content

Benchmark

DSBench-FullStack

Coding benchmarks compare models on programming tasks. Scores are percentages. Higher is better.

Models measured
2
Scores recorded
4
Sources
2
IQ weight
No weight published

Score distribution

69.9%
Mean
71.1%
Median
1.2%
Std dev
68.7%
10th percentile
71.1%
90th percentile
68.7%
Min
68.7%71.1%
The band covers the middle 80% of scores. The line marks the median.

Leaderboard

#ModelScore
1DeepSeek V4 Pro 0813· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim71.1%
2DeepSeek V4 Pro 0813official model-card setting· Reported by DeepSeek· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.71.1%
3DeepSeek V4 Flash 0731· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim68.7%
4DeepSeek V4 Flash 0731official model-card setting· Reported by DeepSeek· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.68.7%

Data credits: BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links. · Data by LLM Stats