jevland

replica source-backed

JevForge: decision-data synthesis, training and web-action scoring

An end-to-end research toolkit builds decision records, trains and calibrates candidate scorers, and serves a Jev-compatible API. Its released Qwen3.5-0.8B model ranks supplied web actions, while the public demo presents recorded Mind2Web decisions.

Recorded candidate decisions and probability labels on JevForge's mock browser surface.
JevForge's recorded Mind2Web candidate replay by zwliJay; API offline during capture. Demo screenshot ↗ ·

Open demo ↗

The model card reports Choice top-1 of 0.5787 on 2,400 test records and 0.6373 on 1,158 OOD records using website-disjoint splits and a saved temperature of 0.9717.

source-backed — Public repository, docs or live artifact. About this label

HTTP availability
Link reachable ·
Method: bounded HTTP request with redirects. HTTP 200 · HTTP response only. Checked destination ↗ · Last reachable
Browser demo scenario
Not checked
Build
Not checked
Live API test
Not checked

What these checks cover →

Report a problem →

← Back to the directory