Coding benchmark
DSBench-FullStack leaderboard
DeepSeek DSBench FullStack. Every model the catalog carries a published DSBench-FullStack value for, ranked by that value.
CategoryCoding
MeasureProvider-reported score
TasksInternal full-stack coding-agent tasks
DifficultyFull-stack software engineering
Published byDeepSeek V4 Flash 0731 update
DeepSeek's internal full-stack coding-agent benchmark.
DSBench-FullStack ranking
2 models with a published DSBench-FullStack value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.
| Rank | Model | Provider | Provider-reported score |
|---|---|---|---|
| 1 |  DeepSeek V4 Pro 0813 | DeepSeek | 71.1 |
| 2 | DeepSeek V4 Flash 0731 | DeepSeek | 68.7 |
Evidence key: Observed
Rows are ordered by the value DeepSeek V4 Flash 0731 update published, highest first. The source does not state whether a higher value is the better result, so this page does not either: for a benchmark that measures a rate of failure, read the table from the bottom.