File size: 1,169 Bytes
f3112f0
ffd3b47
f3112f0
e5f3b98
 
f3112f0
 
 
ffd3b47
e5f3b98
ffd3b47
 
e5f3b98
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1b690c7
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
e5f3b98
 
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
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
---
title: Recipe Lens Chef
sdk: gradio
python_version: "3.12"
suggested_hardware: t4-small
app_file: app.py
---

# Recipe Lens Chef

This repository contains a minimal Hugging Face Space for turning food photos
into chef-style recipe ideas and cooking guidance with
[`openbmb/MiniCPM-V-4.6`](https://huggingface.co/openbmb/MiniCPM-V-4.6).

The app is intentionally inference-only. It does not include fine-tuning,
dataset preparation, trainer scripts, adapters, checkpoints, or training
artifacts.

## Local Development

```bash
uv sync --python 3.12
uv run python app.py
```

The model ID can be overridden without changing code:

```bash
MODEL_ID=openbmb/MiniCPM-V-4.6 uv run python app.py
```

## Space Deployment

Deployments are owned by GitHub Actions. Pushes to `main` run
`.github/workflows/deploy-space.yml`, which uploads the checked-out GitHub
commit to the Hugging Face Space.

Required GitHub secret:

```bash
HF_TOKEN
```

Optional GitHub variable:

```bash
HF_SPACE_ID=binzhango/hf-build-small-hackathon
```

The Space reads dependencies from `requirements.txt` and starts `app.py` with
Gradio. Python 3.12 is requested in this README's Space metadata.