Owner-only columns: policy A/B scores, MCQ format, judged timestamp 30beb3c verified Mushari440 commited on 4 days ago
A requeue must never revive a status that was set on purpose 663c248 Mushari440 Claude Opus 5 commited on 12 days ago
Make the manage panel work across versions, and fix what it did wrong f79409d Mushari440 Claude Opus 5 commited on 12 days ago
Add a benchmark-version picker: v7 keeps the published board, v8 starts empty 055f64d Mushari440 Claude Opus 5 commited on 12 days ago
fix: 500 on delete/rerun — _list_requests never returned _path acb610f verified Mushari440 commited on 20 days ago
fix: manage/private-results 403 — CSRF origin check on read-only GETs e453bf5 verified Mushari440 commited on 20 days ago
feat: private evaluations + maintainer manage panel (delete / re-run) 3571a43 verified Mushari440 commited on 20 days ago
feat: owner-only partial runs (item cap + subtask picker) 702c6fd verified Mushari440 commited on 20 days ago
feat: accept closed-source (API) models, restricted to the maintainer 255c23c verified Mushari440 commited on 20 days ago
Windows mouse-wheel horizontal scroll; remove per-user queue cap 63262c4 verified Mushari440 commited on Jul 14
Add gated model submission: FastAPI backend + HF OAuth sign-in + submit form 3eef8f8 verified Mushari440 commited on Jul 13
Strip gradio app files: clean static Space (index.html only) be63463 verified Mushari440 commited on Jul 13
BISECT TEST: restore app.py @ 0038b5bbfa (known-good multi-row) to isolate code-vs-data 21e377a verified Mushari440 commited on Jul 13
TEST: drop custom theme to check if it broke the table virtualizer 489b175 verified Mushari440 commited on Jul 13
Fix: remove boolean Deleted/incomplete filter (it collapsed the table render to 1 row) f346bc5 verified Mushari440 commited on Jul 13
Fix: stop live-refreshing the custom leaderboard (it collapsed to 1 row); refresh on restart only dbc8df4 verified Mushari440 commited on Jul 13
UI: params slider defaults to 0–72B (was auto-narrowed to the data min/max) d0d524f verified Mushari440 commited on Jul 13
Frontend: Deleted/incomplete filter no longer hides models by default ecf99a7 verified Mushari440 commited on Jul 11
Live auto-refresh leaderboard + queues via gr.Timer (no manual refresh) b2495c3 mushari commited on Jun 24