返回 Pixelle-Video
config-schema.md
根目录 / docs / en / reference / config-schema.md
1 # Config Schema
2
3 Detailed explanation of the `config.yaml` configuration file.
4
5 ---
6
7 ## Configuration Structure
8
9 ```yaml
10 llm:
11 api_key: "your-api-key"
12 base_url: "https://dashscope.aliyuncs.com/compatible-mode/v1"
13 model: "qwen-plus"
14
15 comfyui:
16 comfyui_url: "http://127.0.0.1:8188"
17 comfyui_api_key: "" # ComfyUI API key (optional)
18 runninghub_api_key: ""
19 runninghub_concurrent_limit: 1 # Concurrent limit (1-10)
20 runninghub_instance_type: "" # Instance type (optional, set to "plus" for 48GB VRAM)
21
22 image:
23 default_workflow: "runninghub/image_flux.json"
24 prompt_prefix: "Minimalist illustration style"
25
26 video:
27 default_workflow: "runninghub/video_wan2.1_fusionx.json"
28 prompt_prefix: "Minimalist illustration style"
29
30 tts:
31 default_workflow: "selfhost/tts_edge.json"
32
33 template:
34 default_template: "1080x1920/image_default.html"
35 ```
36
37 ---
38
39 ## LLM Configuration
40
41 - `api_key`: API key
42 - `base_url`: API service address (supports any OpenAI-compatible interface)
43 - `model`: Model name
44
45 ---
46
47 ## ComfyUI Configuration
48
49 ### Basic Configuration
50
51 - `comfyui_url`: Local ComfyUI address (default `http://127.0.0.1:8188`)
52 - `comfyui_api_key`: ComfyUI API key (optional, for [Comfy Platform](https://platform.comfy.org/profile/api-keys))
53
54 ### RunningHub Cloud Configuration
55
56 - `runninghub_api_key`: RunningHub API key (required for cloud workflows)
57 - `runninghub_concurrent_limit`: Concurrent execution limit (1-10, default 1 for regular members)
58 - `runninghub_instance_type`: Instance type (optional)
59 - Empty or unset: Use 24GB VRAM machine
60 - `"plus"`: Use 48GB VRAM machine (suitable for large video generation)
61
62 ### Image Configuration
63
64 - `default_workflow`: Default image generation workflow
65 - `prompt_prefix`: Prompt prefix
66
67 ### Video Configuration
68
69 - `default_workflow`: Default video generation workflow
70 - `runninghub/video_wan2.1_fusionx.json`: Cloud workflow (recommended, no local setup required)
71 - `selfhost/video_wan2.1_fusionx.json`: Local workflow (requires local ComfyUI support)
72 - `prompt_prefix`: Video prompt prefix (controls video generation style)
73
74 ### TTS Configuration
75
76 - `default_workflow`: Default TTS workflow
77
78 ---
79
80 ## Template Configuration
81
82 - `default_template`: Default frame template path (e.g., `1080x1920/image_default.html`)
83
84 ---
85
86 ## More Information
87
88 The configuration file is automatically created on first run.
89
90
90 lines MARKDOWN