AyhamJo7's picture
feat(agent): implement GAIA evaluation workflow
6f718f1
|
Raw
History Blame Contribute Delete
808 Bytes

A newer version of the Gradio SDK is available: 6.21.0

Upgrade
metadata
title: GAIA Final Agent
emoji: 🎯
colorFrom: blue
colorTo: green
sdk: gradio
sdk_version: 5.25.2
app_file: app.py
pinned: false
hf_oauth: true
hf_oauth_expiration_minutes: 480

GAIA Final Agent

Tool-using agent for the Hugging Face Agents Course final assignment. It supports web research, webpage retrieval, spreadsheets, PDFs, trusted benchmark Python files, audio transcription, YouTube transcription, and exact-answer normalization.

The submitted evaluation for AyhamJo7 scored 100% (20/20).

Local development

uv sync --extra dev
uv run gaia-agent evaluate --output answers.json
uv run gaia-agent submit --answers answers.json

Local execution uses Ollama at http://127.0.0.1:11434. The hosted Space uses Hugging Face inference through the HF_TOKEN Space secret.