agents benchmark
CL-bench Leaderboard
CL-bench is an open-source benchmark for evaluating models on coding and agentic tasks using its own data and rubrics.
Updated Sep 5, 2026
Models2
Model coverage2
MetricScore
EvidenceB
CL-bench Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelTE | Score23.80% | Percentile100.00% | Participants2 | EvidenceC | Evaluated |
| Rank02 | ModelMI | Score20.48% | Percentile0.00% | Participants2 | EvidenceC | Evaluated |
CL-bench Highlights
The leading models and scores on this benchmark.
CL-bench Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for CL-bench
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is CL-bench?
What CL-bench measures and how its scores work.
CL-bench is an open-source benchmark with its own data and rubrics for evaluating models on coding and agentic tasks.
It measures performance on coding and agentic tasks using Score, reported as a ratio.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- CL-bench
- Modality
- text
- Primary category
- agents
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- overall
LLM Stats. Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about CL-bench.
Which model scores highest on CL-bench?
Hy3 is currently ranked first with 23.80%.
What are the top three models on CL-bench?
The current leaders are Hy3 (23.80%) and MiniMax M3 (20.48%).
Which CL-bench model has the lowest official input price?
MiniMax M3 has the lowest matched official input price at $0.30 input / $1.2 output per 1M tokens.
Which models are fastest among CL-bench results?
The fastest matched records are MiniMax M3 (6.61 tok/s via MiniMax).
Does the highest score result prove overall model quality?
No. This benchmark measures one defined capability or task. The overall LLMBoard score uses a separate aggregation across eligible benchmark evidence.