kaust-generative-ai/telco-gaia-leaderboard

Telco-GAIA Leaderboard

A static leaderboard for the Telco-GAIA benchmark: the reference agent's full sweep across commercial and open LLMs on all 100 tasks (single run), sorted by accuracy — with search, sortable columns, and column toggles. Duration and turns are per task.

† run under rate limits, so durations may be slightly inflated.

Updating the numbers

Edit data.json (one record per model) and re-upload it — index.html fetches it at load time, so no code changes are needed to add rows.