DeepSeek DSBench Hard
DeepSeek's internal hard coding-agent benchmark.
2 models
Higher is better.
About this benchmark
- Metric
- Provider-reported score
- Tasks
- Internal hard coding-agent tasks
- Authors
- DeepSeek-AI (2026)
Scores are as reported by model developers or evaluators and may use different settings (for example reasoning effort or tools). Compare with care.