multimodal benchmark
AndroidWorld_SR Leaderboard
AndroidWorld Success Rate (SR) benchmark - A dynamic benchmarking environment for autonomous agents operating on Android devices. Evaluates agents on 116 programmatic tasks across 20 real-world Android apps using multimodal inputs (screen screenshots, accessibility trees, and natural language instructions). Measures success rate of agents completing tasks like sending messages, creating calendar events, and navigating mobile interfaces. Published at ICLR 2025. Best current performance: 30.6% success rate (M3A agent) vs 80.0% human performance.
Updated Aug 17, 2026
Models8
Model coverage8
MetricScore
EvidenceB
AndroidWorld_SR Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelAC | Score71.1% | Percentile100.0% | Participants8 | EvidenceC | Evaluated |
| Rank02 | ModelAC | Score66.4% | Percentile85.7% | Participants8 | EvidenceC | Evaluated |
| Rank03 | ModelAC | Score64.2% | Percentile71.4% | Participants8 | EvidenceC | Evaluated |
| Rank04 | ModelAC | Score63.7% | Percentile57.1% | Participants8 | EvidenceC | Evaluated |
| Rank05 | ModelAC | Score63.7% | Percentile42.9% | Participants8 | EvidenceC | Evaluated |
| Rank06 | ModelAC | Score35.0% | Percentile28.6% | Participants8 | EvidenceC | Evaluated |
| Rank07 | ModelAC | Score25.5% | Percentile14.3% | Participants8 | EvidenceC | Evaluated |
| Rank08 | ModelAC | Score22.0% | Percentile0.0% | Participants8 | EvidenceC | Evaluated |
AndroidWorld_SR Highlights
The leading models and scores on this benchmark.
AndroidWorld_SR Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for AndroidWorld_SR
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
Ranking basisThis androidworld_sr 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.
- 01ACQwen3.5-35B-A3BAlibaba Cloud / Qwen Team
- Score
- 71.1%
- Price
- $0.25 input / $2.0 output per 1M tokens
Strengths
- Ranks #1 of 8 compared models
- 100th percentile on this benchmark
- C evidence result
Considerations
- This result measures AndroidWorld_SR, not total model capability
- 02ACQwen3.5-122B-A10BAlibaba Cloud / Qwen Team
- Score
- 66.4%
- Price
- $0.40 input / $3.2 output per 1M tokens
Strengths
- Ranks #2 of 8 compared models
- 86th percentile on this benchmark
- C evidence result
Considerations
- This result measures AndroidWorld_SR, not total model capability
- 03ACQwen3.5-27BAlibaba Cloud / Qwen Team
- Score
- 64.2%
- Price
- $0.30 input / $2.4 output per 1M tokens
- Speed
- Up to 6.7 tok/s via Novita
Strengths
- Ranks #3 of 8 compared models
- 71th percentile on this benchmark
- C evidence result
Considerations
- This result measures AndroidWorld_SR, not total model capability
- 04ACQwen3 VL 235B A22B InstructAlibaba Cloud / Qwen Team
- Score
- 63.7%
Strengths
- Ranks #4 of 8 compared models
- 57th percentile on this benchmark
- C evidence result
Considerations
- This result measures AndroidWorld_SR, not total model capability
- 05ACQwen3 VL 32B ThinkingAlibaba Cloud / Qwen Team
- Score
- 63.7%
Strengths
- Ranks #5 of 8 compared models
- 43th percentile on this benchmark
- C evidence result
Considerations
- This result measures AndroidWorld_SR, not total model capability
Selection summary
Best AI Models for AndroidWorld_SR
Qwen3.5-35B-A3B currently leads AndroidWorld_SR with 71.1%. 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 AndroidWorld_SR?
What AndroidWorld_SR measures and how its scores work.
AndroidWorld Success Rate (SR) benchmark - A dynamic benchmarking environment for autonomous agents operating on Android devices. Evaluates agents on 116 programmatic tasks across 20 real-world Android apps using multimodal inputs (screen screenshots, accessibility trees, and natural language instructions). Measures success rate of agents completing tasks like sending messages, creating calendar events, and navigating mobile interfaces. Published at ICLR 2025. Best current performance: 30.6% success rate (M3A agent) vs 80.0% human performance.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- AndroidWorld_SR
- Modality
- multimodal
- Primary category
- multimodal
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- androidworld-sr|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about AndroidWorld_SR.
Which model scores highest on AndroidWorld_SR?
Qwen3.5-35B-A3B is currently ranked first with 71.1%.
What does AndroidWorld_SR measure?
AndroidWorld Success Rate (SR) benchmark - A dynamic benchmarking environment for autonomous agents operating on Android devices. Evaluates agents on 116 programmatic tasks across 20 real-world Android apps using multimodal inputs (screen screenshots, accessibility trees, and natural language instructions). Measures success rate of agents completing tasks like sending messages, creating calendar events, and navigating mobile interfaces. Published at ICLR 2025. Best current performance: 30.6% success rate (M3A agent) vs 80.0% human performance.
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.