AndroidBench
AndroidBench - evaluated on the 95-task public subset, reporting avg@3 scores.
Category
Agentic Coding
Max Score
—
Score Type
percent
Active
Yes
Model Scores
Scores for AndroidBench
| Model | Score | Date | Verified |
|---|---|---|---|
| Qwen3.8-2.4T-A95B |
66.43%
|
18.08.2026 | Verified |
| Fable-5 |
100.00%
|
18.08.2026 | Unverified |
| GPT-5.6 Sol |
62.50%
|
18.08.2026 | Unverified |
| Claude Opus 4.8 |
47.50%
|
18.08.2026 | Unverified |
| Qwen3.7-Max |
56.50
|
18.08.2026 | Unverified |