Unspecified benchmark
LM Arena Agent Task Outcome Explicit Leaderboard
LM Arena Agent Task Outcome Explicit is a current leaderboard using the Agent Score metric, reported as a ratio.
Updated Sep 8, 2026
Models49
Model coverage49
MetricAgent Score
EvidenceA
LM Arena Agent Task Outcome Explicit Ranking
Higher agent score ranks better on this benchmark.
Sort by
Rank | Model | Agent Score | Percentile | Participants | Observations | Evidence | Evaluated |
|---|
| Rank01 | ModelAN | Agent Score22.45% (18.73% - 26.16%) | Percentile100.00% | Participants49 | Observations2.5K | EvidenceA | Evaluated |
| Rank02 | ModelMA | Agent Score16.58% (15.19% - 17.98%) | Percentile97.92% | Participants49 | Observations74.2K | EvidenceA | Evaluated |
| Rank03 | ModelAN | Agent Score14.89% (10.95% - 18.84%) | Percentile95.83% | Participants49 | Observations12.5K | EvidenceA | Evaluated |
| Rank04 | ModelZA | Agent Score14.14% (11.98% - 16.30%) | Percentile93.75% | Participants49 | Observations13.5K | EvidenceA | Evaluated |
| Rank05 | ModelTE | Agent Score13.56% (10.60% - 16.52%) | Percentile91.67% | Participants49 | Observations6.5K | EvidenceA | Evaluated |
| Rank06 | ModelAC | Agent Score12.46% (9.97% - 14.96%) | Percentile89.58% | Participants49 | Observations9.5K | EvidenceA | Evaluated |
| Rank07 | ModelZA | Agent Score12.10% (10.57% - 13.62%) | Percentile87.50% | Participants49 | Observations34.9K | EvidenceA | Evaluated |
| Rank08 | ModelDE | Agent Score11.75% (9.84% - 13.66%) | Percentile85.42% | Participants49 | Observations20.4K | EvidenceA | Evaluated |
| Rank09 | ModelGO | Agent Score10.42% (6.01% - 14.84%) | Percentile83.33% | Participants49 | Observations4.6K | EvidenceA | Evaluated |
| Rank10 | ModelAC | Agent Score10.37% (8.02% - 12.73%) | Percentile81.25% | Participants49 | Observations14K | EvidenceA | Evaluated |
| Rank11 | ModelGO | Agent Score9.54% (7.38% - 11.71%) | Percentile79.17% | Participants49 | Observations19.6K | EvidenceA | Evaluated |
| Rank12 | ModelZA | Agent Score8.63% (6.98% - 10.27%) | Percentile77.08% | Participants49 | Observations42.5K | EvidenceA | Evaluated |
| Rank13 | ModelDE | Agent Score8.46% (6.69% - 10.22%) | Percentile75.00% | Participants49 | Observations35.4K | EvidenceA | Evaluated |
| Rank14 | ModelXA | Agent Score7.83% (5.08% - 10.59%) | Percentile72.92% | Participants49 | Observations12.9K | EvidenceA | Evaluated |
| Rank15 | ModelAN | Agent Score7.37% (4.20% - 10.54%) | Percentile70.83% | Participants49 | Observations27.5K | EvidenceA | Evaluated |
| Rank16 | ModelAC | Agent Score6.98% (4.72% - 9.24%) | Percentile68.75% | Participants49 | Observations13.9K | EvidenceA | Evaluated |
| Rank17 | ModelOP | Agent Score6.68% (3.55% - 9.81%) | Percentile66.67% | Participants49 | Observations22.8K | EvidenceA | Evaluated |
| Rank18 | ModelME | Agent Score6.60% (3.78% - 9.41%) | Percentile64.58% | Participants49 | Observations12.9K | EvidenceA | Evaluated |
| Rank19 | ModelXA | Agent Score6.30% (3.75% - 8.86%) | Percentile62.50% | Participants49 | Observations23.8K | EvidenceA | Evaluated |
| Rank20 | ModelAN | Agent Score5.93% (2.49% - 9.36%) | Percentile60.42% | Participants49 | Observations21.1K | EvidenceA | Evaluated |
| Rank21 | ModelME | Agent Score5.01% (3.47% - 6.55%) | Percentile58.33% | Participants49 | Observations56.5K | EvidenceA | Evaluated |
| Rank22 | ModelAN | Agent Score3.51% (0.36% - 6.67%) | Percentile56.25% | Participants49 | Observations25.8K | EvidenceA | Evaluated |
| Rank23 | ModelAN | Agent Score2.80% (-1.33% - 6.93%) | Percentile54.17% | Participants49 | Observations19.4K | EvidenceA | Evaluated |
| Rank24 | ModelAN | Agent Score2.50% (-0.67% - 5.66%) | Percentile52.08% | Participants49 | Observations27.8K | EvidenceA | Evaluated |
| Rank25 | ModelOP | Agent Score2.07% (-0.35% - 4.49%) | Percentile50.00% | Participants49 | Observations39.1K | EvidenceA | Evaluated |
| Rank26 | ModelMA | Agent Score1.15% (-4.77% - 7.06%) | Percentile47.92% | Participants49 | Observations5.7K | EvidenceA | Evaluated |
| Rank27 | ModelOP | Agent Score1.14% (-1.00% - 3.27%) | Percentile45.83% | Participants49 | Observations53.9K | EvidenceA | Evaluated |
| Rank28 | ModelOP | Agent Score0.83% (-1.81% - 3.46%) | Percentile43.75% | Participants49 | Observations15K | EvidenceA | Evaluated |
| Rank29 | ModelZA | Agent Score0.17% (-1.98% - 2.32%) | Percentile41.67% | Participants49 | Observations45.3K | EvidenceA | Evaluated |
| Rank30 | ModelGO | Agent Score-1.62% (-3.60% - 0.37%) | Percentile39.58% | Participants49 | Observations68.3K | EvidenceA | Evaluated |
LM Arena Agent Task Outcome Explicit Highlights
The leading models and scores on this benchmark.
LM Arena Agent Task Outcome Explicit Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for LM Arena Agent Task Outcome Explicit
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is LM Arena Agent Task Outcome Explicit?
What LM Arena Agent Task Outcome Explicit measures and how its scores work.
LM Arena Agent Task Outcome Explicit is a current leaderboard.
It reports Agent Score as a ratio.
Scores are shown in ratio. This benchmark is verified and has an evidence level of A.
- Family
- lm-arena-agent-task-outcome-explicit
- Modality
- Unspecified
- Primary category
- Unspecified
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- overall
LM Arena. Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about LM Arena Agent Task Outcome Explicit.
Which model scores highest on LM Arena Agent Task Outcome Explicit?
Claude Fable 5.1 is currently ranked first with 22.45%.
What are the top three models on LM Arena Agent Task Outcome Explicit?
The current leaders are Claude Fable 5.1 (22.45%), Kimi K3 (16.58%), and Claude Opus 5 (14.89%).
Which LM Arena Agent Task Outcome Explicit model has the lowest official input price?
GLM-5.3-Flash has the lowest matched official input price at $0.08 input / $0.25 output per 1M tokens.
Which models are fastest among LM Arena Agent Task Outcome Explicit results?
The fastest matched records are GLM-5.3 (472.69 tok/s via FriendliAI), Gemini 3.5 Flash (210.94 tok/s via Google), and GPT-5.5 (134.94 tok/s via OpenAI).
Does the highest agent 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.