Android Bench
AI-assisted software engineering has seen the emergence of several benchmarks to measure the capabilities of LLMs. Android developers face specific challenges that aren't covered by existing benchmarks, so we created one that focuses on a north star of high quality Android development.
Android LLM Leaderboard
| Model | Score (%) |
arrow_range Cl range (%) |
Date |
|---|---|---|---|
|
72.4 | 65.3 — 79.8 | 2026-03-04 |
|
66.6 | 58.9 — 73.9 | 2026-03-04 |
|
62.5 | 54.7 — 70.3 | 2026-03-04 |
|
|
61.9 | 53.9 — 69.6 | 2026-03-04 |
|
|
60.4 | 52.6 — 67.8 | 2026-03-04 |
|
|
58.4 | 51.1 — 66.6 | 2026-03-04 |
|
|
54.2 | 45.5 — 62.4 | 2026-03-04 |
|
|
42.0 | 36.3 — 47.9 | 2026-03-04 |
|
|
16.1 | 10.9 — 21.9 | 2026-03-04 |
Latest results as of March 5th 2026 - check back periodically for updates!
Score is the average percentage of 100 test cases successfully resolved across 10 runs for each model.
Confidence Interval (CI) represents the expected performance range, reflecting the results' statistical reliability (p-value < 0.05).
