agents benchmark
Claw-Eval Leaderboard
Claw-Eval tests real-world agentic task completion across complex multi-step scenarios, evaluating a model's ability to use tools, navigate environments, and complete end-to-end tasks autonomously.
Updated Aug 17, 2026
Models14
Model coverage14
MetricScore
EvidenceB
Claw-Eval Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelMA | Score80.9% | Percentile100.0% | Participants14 | EvidenceC | Evaluated |
| Rank02 | ModelZA | Score75.0% | Percentile92.3% | Participants14 | EvidenceC | Evaluated |
| Rank03 | ModelMI | Score74.5% | Percentile84.6% | Participants14 | EvidenceC | Evaluated |
| Rank04 | ModelTE | Score68.5% | Percentile76.9% | Participants14 | EvidenceC | Evaluated |
| Rank05 | ModelAC | Score65.2% | Percentile69.2% | Participants14 | EvidenceC | Evaluated |
| Rank06 | ModelXI | Score64.0% | Percentile61.5% | Participants14 | EvidenceC | Evaluated |
| Rank07 | ModelXI | Score63.2% | Percentile53.9% | Participants14 | EvidenceC | Evaluated |
| Rank08 | ModelLA | Score62.8% | Percentile46.1% | Participants14 | EvidenceC | Evaluated |
| Rank09 | ModelAC | Score62.7% | Percentile38.5% | Participants14 | EvidenceC | Evaluated |
| Rank10 | ModelXI | Score61.5% | Percentile30.8% | Participants14 | EvidenceC | Evaluated |
| Rank11 | ModelAC | Score60.6% | Percentile23.1% | Participants14 | EvidenceC | Evaluated |
| Rank12 | ModelAC | Score58.7% | Percentile15.4% | Participants14 | EvidenceC | Evaluated |
| Rank13 | ModelXI | Score54.8% | Percentile7.7% | Participants14 | EvidenceC | Evaluated |
| Rank14 | ModelAC | Score50.0% | Percentile0.0% | Participants14 | EvidenceC | Evaluated |
Claw-Eval Highlights
The leading models and scores on this benchmark.
Claw-Eval Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for Claw-Eval
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
Ranking basisThis claw-eval 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.
- 01MAKimi K2.6Moonshot AI
- Score
- 80.9%
- Price
- $0.95 input / $4.0 output per 1M tokens
- Speed
- Up to 285 tok/s via Fireworks
Strengths
- Ranks #1 of 14 compared models
- 100th percentile on this benchmark
- C evidence result
Considerations
- This result measures Claw-Eval, not total model capability
- 02ZAGLM-5V-TurboZhipu AI
- Score
- 75.0%
- Price
- $5.0 input / $22 output per 1M tokens
Strengths
- Ranks #2 of 14 compared models
- 92th percentile on this benchmark
- C evidence result
Considerations
- This result measures Claw-Eval, not total model capability
- 03MIMiniMax M3MiniMax
- Score
- 74.5%
- Price
- $0.30 input / $1.2 output per 1M tokens
- Speed
- Up to 214 tok/s via Together
Strengths
- Ranks #3 of 14 compared models
- 85th percentile on this benchmark
- C evidence result
Considerations
- This result measures Claw-Eval, not total model capability
- 04TEHy3Tencent
- Score
- 68.5%
Strengths
- Ranks #4 of 14 compared models
- 77th percentile on this benchmark
- C evidence result
Considerations
- This result measures Claw-Eval, not total model capability
- 05ACQwen3.7 MaxAlibaba Cloud / Qwen Team
- Score
- 65.2%
- Price
- $2.5 input / $7.5 output per 1M tokens
- Speed
- Up to 5.8 tok/s via Together
Strengths
- Ranks #5 of 14 compared models
- 69th percentile on this benchmark
- C evidence result
Considerations
- This result measures Claw-Eval, not total model capability
Selection summary
Best AI Models for Claw-Eval
Kimi K2.6 currently leads Claw-Eval with 80.9%. 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 Claw-Eval?
What Claw-Eval measures and how its scores work.
Claw-Eval tests real-world agentic task completion across complex multi-step scenarios, evaluating a model's ability to use tools, navigate environments, and complete end-to-end tasks autonomously.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- Claw-Eval
- Modality
- text
- Primary category
- agents
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- claw-eval|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about Claw-Eval.
Which model scores highest on Claw-Eval?
Kimi K2.6 is currently ranked first with 80.9%.
What does Claw-Eval measure?
Claw-Eval tests real-world agentic task completion across complex multi-step scenarios, evaluating a model's ability to use tools, navigate environments, and complete end-to-end tasks autonomously.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
14 model results are currently shown.
Does this benchmark affect the overall score?
Yes. This benchmark can contribute to the current LLMBoard capability score.