agents benchmark
DSBench-Hard Leaderboard
DSBench-Hard is DeepSeek's internal test set of difficult coding-agent problems.
Updated Aug 17, 2026
Models2
Model coverage2
MetricScore
EvidenceB
DSBench-Hard Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelDE | Score67.2% | Percentile100.0% | Participants2 | EvidenceC | Evaluated |
| Rank02 | ModelDE | Score59.6% | Percentile0.0% | Participants2 | EvidenceC | Evaluated |
DSBench-Hard Highlights
The leading models and scores on this benchmark.
DSBench-Hard Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for DSBench-Hard
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
Ranking basisThis dsbench-hard AI model leaderboard uses descending score in the benchmark's original unit. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.
- 01DEDeepSeek-V4-Pro-0813DeepSeek
- Score
- 67.2%
- Price
- $0.43 input / $0.87 output per 1M tokens
- Speed
- Up to 33 tok/s via DeepSeek
Strengths
- Ranks #1 of 2 compared models
- 100th percentile on this benchmark
- C evidence result
Considerations
- This result measures DSBench-Hard, not total model capability
- 02DEDeepSeek-V4-Flash-0731DeepSeek
- Score
- 59.6%
- Price
- $0.14 input / $0.28 output per 1M tokens
- Speed
- Up to 19 tok/s via Novita
Strengths
- Ranks #2 of 2 compared models
- 0th percentile on this benchmark
- C evidence result
Considerations
- This result measures DSBench-Hard, not total model capability
Selection summary
Best AI Models for DSBench-Hard
DeepSeek-V4-Pro-0813 currently leads DSBench-Hard with 67.2%. It is the top model on this specific benchmark, while the best LLM for the broader task should also be checked against other benchmarks, price and runtime.
Use this leaderboard with the supporting benchmark results and coverage details above. A leaderboard position summarizes the selected ranking signal; it does not replace workload-specific testing.
What is DSBench-Hard?
What DSBench-Hard measures and how its scores work.
DSBench-Hard is DeepSeek's internal test set of difficult coding-agent problems.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- DSBench-Hard
- Modality
- text
- Primary category
- agents
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- dsbench-hard|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about DSBench-Hard.
Which model scores highest on DSBench-Hard?
DeepSeek-V4-Pro-0813 is currently ranked first with 67.2%.
What does DSBench-Hard measure?
DSBench-Hard is DeepSeek's internal test set of difficult coding-agent problems.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
2 model results are currently shown.
Does this benchmark affect the overall score?
No. This benchmark is shown for reference but does not contribute to the overall score.