Spaces:
Running on Zero
Running on Zero
File size: 1,107 Bytes
2705160 62ad9da 2705160 d123508 2705160 3e9095a | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 | ---
title: Template Final Assignment
emoji: 🕵🏻♂️
colorFrom: indigo
colorTo: indigo
sdk: gradio
sdk_version: 5.25.2
app_file: app.py
pinned: false
hf_oauth: true
# optional, default duration is 8 hours/480 minutes. Max duration is 30 days/43200 minutes.
hf_oauth_expiration_minutes: 480
---
Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
## Agent architecture
- The manager uses `GROQ_MANAGER_MODEL` (default: `openai/gpt-oss-120b`).
- Web research is one direct, bounded Tavily request; it does not start a
second LLM agent.
- Document and media tasks keep dedicated agents because they may need to
download and inspect an attachment in multiple steps.
- Calculations use `python_exec` or `calculator` directly.
- Final answer formatting uses `GROQ_NORMALIZER_MODEL` (default:
`openai/gpt-oss-20b`).
Required Space secrets are `GROQ_API_KEY` and `TAVILY_API_KEY`. Set
`GROQ_VLM_MODEL` as well to enable image and video analysis.
Run the offline web-search regression tests with:
```bash
python -m unittest -v test_web_search.py
```
|