Spaces:
Sleeping
Sleeping
LeVinh commited on
Commit ·
2f8f248
1
Parent(s): 2609185
change readme to run submission
Browse files
README.md
CHANGED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
title: GAIA Agent Submission
|
| 3 |
+
emoji: 🚀
|
| 4 |
+
colorFrom: indigo
|
| 5 |
+
colorTo: blue
|
| 6 |
+
sdk: docker
|
| 7 |
+
app_port: 7860
|
| 8 |
+
app_file: eval.py
|
| 9 |
+
pinned: false
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
# 🤖 GAIA Agent Submission Runner
|
| 13 |
+
|
| 14 |
+
This Space is configured to run the **GAIA Benchmark Evaluation & Submission**.
|
| 15 |
+
|
| 16 |
+
## 📋 How to Submit
|
| 17 |
+
|
| 18 |
+
1. **Login**: Click the "Login using Hugging Face" button in the app.
|
| 19 |
+
2. **Submit**: Click "Run Evaluation & Submit All Answers".
|
| 20 |
+
* The agent will fetch questions, solve them, and auto-submit the results to the leaderboard.
|
| 21 |
+
|
| 22 |
+
## ⚙️ Configuration
|
| 23 |
+
|
| 24 |
+
This agent uses a generic OpenAI-compatible provider (configured for SambaNova by default) via `agent.json`.
|
| 25 |
+
|
| 26 |
+
**Required Secrets:**
|
| 27 |
+
Ensure these are set in your Space Settings:
|
| 28 |
+
* `SAMBANOVA_API_KEY`: Your SambaNova API key (or equivalent for your chosen provider).
|
| 29 |
+
* `HF_TOKEN`: (Optional but recommended) For accessing private datasets if needed.
|
| 30 |
+
|
| 31 |
+
## 🛠️ Local Development
|
| 32 |
+
|
| 33 |
+
* **Chat Demo**: Run `python app.py` locally to use the interactive chat interface.
|
| 34 |
+
* **Evaluation**: Run `python eval.py` locally to test the submission logic.
|