agents benchmark
SkillsBench
SkillsBench evaluates coding agents on self-contained programming tasks, measuring practical engineering skills across diverse software development scenarios.
Updated Aug 11, 2026
Models8
Model coverage8
MetricScore
EvidenceB
SkillsBench Ranking
Higher score ranks better on this benchmark.
| 1 | AC | 70.2% | 100.0% | 8 | C | |
| 2 | AC | 59.2% | 85.7% | 8 | C | |
| 3 | TE | 55.3% | 71.4% | 8 | C | |
| 4 | AC | 54.9% | 57.1% | 8 | C | |
| 5 | AC | 48.2% | 42.9% | 8 | C | |
| 6 | AC | 45.7% | 28.6% | 8 | C | |
| 7 | ME | 44.3% | 14.3% | 8 | C | |
| 8 | AC | 28.7% | 0.0% | 8 | C |
SkillsBench Score Distribution
A closer view of the leading scores on this benchmark.
SkillsBench Highlights
The leading models and scores on this benchmark.
What is SkillsBench?
What SkillsBench measures and how its scores work.
SkillsBench evaluates coding agents on self-contained programming tasks, measuring practical engineering skills across diverse software development scenarios.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- SkillsBench
- Modality
- text
- Primary category
- agents
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- skillsbench|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about SkillsBench.
Which model scores highest on SkillsBench?
Qwen3.8 Max is currently ranked first with 70.2%.
What does SkillsBench measure?
SkillsBench evaluates coding agents on self-contained programming tasks, measuring practical engineering skills across diverse software development scenarios.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
8 model results are currently shown.
Does this benchmark affect the overall score?
Yes. This benchmark can contribute to the current LLMBoard capability score.