MiniWeb Full369 - Verify: leaderboard
MiniWeb v1 contains 369 human-authored tasks, 37 primitives, and 65 sites. The main 10-agent study uses Gemini 3.5 Flash with a majority of 3 judge votes. Source interfaces comprise 4 general VLMs using browser-use and 6 native agent loops. The training study uses a Qwen3.5-4B planner and rank-64 LoRA executor, with 27,646 skill steps or 27,644 task-centric steps. Decoding is greedy. MiniWeb and WebArena-Lite use 3 runs; live WebVoyager and OnlineMind2Web use 1 run on normalized task sets of 536 and 280 tasks. WA-Lite has 154 tasks and uses a Gemini 3.5 Flash-Lite fuzzy judge for 43 cases. The planner uses browser-use 0.13.3 flash mode, v6, with at most 25 turns. Exact harnesses and trained weights are retained. Skill families score their own primitive instances; operation columns use their own operation denominators. Metric: Success (%).
Metric: Success (%). Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 10 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | MiniWeb Fara1.5-4B native loop | 76 | |
| 2 | MiniWeb Qwen3.5-9B browser-use | 71 | |
| 3 | MiniWeb Qwen3-VL-8B browser-use | 62 | |
| 4 | MiniWeb MolmoWeb-8B native loop | 52 | |
| 5 | MiniWeb OpenCUA-7B native loop | 52 | |
| 6 | MiniWeb InternVL3.5-8B browser-use | 48 | |
| 7 | MiniWeb Qwen3.5-4B browser-use | 43 | |
| 8 | MiniWeb MolmoWeb-4B native loop | 38 | |
| 9 | MiniWeb UI-TARS-1.5-7B native loop | 33 | |
| 10 | MiniWeb Fara-7B native loop | 29 |
Interactive version: theaggregate.ai/benchmark?slug=miniweb-full369-verify · How It Works · Data refreshed daily, snapshot 2026-10-28.