Skip to content

Commit e6cd670

Browse files
lzx1413lzx1413
committed
feat(docs): add MkDocs + Material documentation site for GitHub Pages
- Create mkdocs.yml with Material theme, i18n plugin (en/zh), CJK slugify - Add bilingual index pages (en/index.md, zh/index.md) - Add GitHub Actions workflow for auto-deployment to Pages - Add mkdocs-static-i18n dependency to pyproject.toml - Fix broken links: _zh.md suffix references, cross-language paths, CJK anchors - Add language switcher (extra.alternate) for en/zh navigation
1 parent c2377e0 commit e6cd670

13 files changed

Lines changed: 774 additions & 9 deletions

File tree

‎.github/workflows/docs.yml‎

Lines changed: 62 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,62 @@
1+
name: Deploy Docs to GitHub Pages
2+
3+
on:
4+
push:
5+
branches: [main]
6+
paths:
7+
- "docs/**"
8+
- "mkdocs.yml"
9+
# Allow manual trigger
10+
workflow_dispatch:
11+
12+
# Sets permissions of the GITHUB_TOKEN to allow deployment to GitHub Pages
13+
permissions:
14+
contents: read
15+
pages: write
16+
id-token: write
17+
18+
# Allow only one concurrent deployment, skipping runs queued between the run in-progress and latest queued.
19+
# However, do NOT cancel in-progress runs as we want to allow these production deployments to complete.
20+
concurrency:
21+
group: "pages"
22+
cancel-in-progress: false
23+
24+
jobs:
25+
# Build job
26+
build:
27+
runs-on: ubuntu-latest
28+
steps:
29+
- name: Checkout
30+
uses: actions/checkout@v4
31+
with:
32+
fetch-depth: 0
33+
34+
- name: Setup Python
35+
uses: actions/setup-python@v5
36+
with:
37+
python-version: "3.12"
38+
39+
- name: Install dependencies
40+
run: |
41+
pip install --upgrade pip
42+
pip install ".[docs]"
43+
44+
- name: Build MkDocs site
45+
run: mkdocs build --strict
46+
47+
- name: Upload artifact
48+
uses: actions/upload-pages-artifact@v3
49+
with:
50+
path: ./site
51+
52+
# Deployment job
53+
deploy:
54+
environment:
55+
name: github-pages
56+
url: ${{ steps.deployment.outputs.page_url }}
57+
runs-on: ubuntu-latest
58+
needs: build
59+
steps:
60+
- name: Deploy to GitHub Pages
61+
id: deployment
62+
uses: actions/deploy-pages@v4

‎docs/en/index.md‎

Lines changed: 66 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,66 @@
1+
# TeleFuser
2+
3+
A **high-performance runtime** for world model inference and multimodal generation.
4+
5+
## Features
6+
7+
- 🌍 **World Model Runtime** — Continuous execution, stateful sessions, bidirectional control loops
8+
- 🚀 **High Performance** — Optimized Triton kernels, feature caching, and multi-GPU inference
9+
- 🎨 **Multimodal Generation** — Image/video generation, super-resolution, speech-to-video
10+
- 📡 **Streaming Transport** — WebRTC with media tracks plus DataChannel for real-time inference
11+
- 🔧 **Flexible Configuration** — Attention implementations, parallel strategies, quantization, offloading
12+
- 📦 **Extensible** — Easy to add new models, stages, and pipelines
13+
14+
## Supported Models
15+
16+
### World Model and Real-Time
17+
18+
| Model | Tasks | Description |
19+
|-------|-------|-------------|
20+
| LingBot-World-Fast | Bidirectional streaming | Interactive world model via WebRTC DataChannel |
21+
22+
### Video Generation
23+
24+
| Model | Tasks | Description |
25+
|-------|-------|-------------|
26+
| WanVideo (Wan2.1/2.2) | T2V, I2V, FL2V | Video generation and editing |
27+
| HunyuanVideo | T2V, I2V | Video generation |
28+
| LTX Video | I2V + Audio | Video generation with audio |
29+
| FlashVSR | VSR | Video super-resolution |
30+
| LiveAct | S2V | Speech-to-video |
31+
| LongCat-Video | T2V, I2V | Long video generation |
32+
33+
### Image Generation
34+
35+
| Model | Tasks | Description |
36+
|-------|-------|-------------|
37+
| Qwen-Image | T2I, Edit | Image generation and editing |
38+
| Z-Image | T2I | Image generation |
39+
| Flux2 Klein | T2I | Image generation |
40+
41+
## Quick Start
42+
43+
```bash
44+
# Install
45+
pip install telefuser
46+
47+
# Batch serving
48+
telefuser serve /path/to/pipeline.py --port 8000
49+
50+
# Stream serving (requires WebRTC)
51+
pip install -e ".[webrtc]"
52+
telefuser stream-serve examples/stream_server/stream_lingbot_world_fast.py -p 8088
53+
```
54+
55+
## Documentation Sections
56+
57+
- **[Service Guide](service.md)** — Batch serving, task APIs, and SDK
58+
- **[Stream Server](stream_server.md)** — WebRTC streaming and bidirectional control
59+
- **[Configuration](configuration.md)** — Runtime and model configuration
60+
- **[Parallel Inference](parallel.md)** — Distributed processing strategies
61+
- **[Adding New Model](adding_new_model.md)** — Integrate new models
62+
- **[Profiler](profiler.md)** — Performance analysis tools
63+
64+
---
65+
66+
[Switch to Chinese 🇨🇳](/zh/)

‎docs/en/service.md‎

Lines changed: 62 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1,11 +1,13 @@
11
# TeleFuser Service Guide
22

3-
This guide covers the TeleFuser API server, CLI usage, and HTTP API reference.
3+
This guide covers the TeleFuser API server, CLI usage, and HTTP API reference. For real-time streaming via WebRTC, see the [Stream Server Guide](stream_server.md).
44

55
## Table of Contents
66

77
- [Quick Start](#quick-start)
8+
- [Serving Modes](#serving-modes)
89
- [CLI Usage](#cli-usage)
10+
- [Supported Pipelines](#supported-pipelines)
911
- [Server Configuration](#server-configuration)
1012
- [Service Metadata Guide](./service_metadata.md)
1113
- [HTTP API Reference](#http-api-reference)
@@ -40,6 +42,11 @@ telefuser serve \
4042
--task t2i \
4143
--port 8000 \
4244
--parallelism 1
45+
46+
# For real-time world model streaming (requires WebRTC support)
47+
pip install -e ".[webrtc]"
48+
export LINGBOT_WORLD_CHECKPOINT_DIR=/path/to/LingBot-World
49+
telefuser stream-serve examples/stream_server/stream_lingbot_world_fast.py -p 8088 --skip-validation
4350
```
4451

4552
### 3. Create a Task
@@ -57,6 +64,60 @@ curl -X POST "http://127.0.0.1:8000/v1/tasks/create" \
5764

5865
---
5966

67+
## Serving Modes
68+
69+
TeleFuser provides two serving commands optimized for different workload types:
70+
71+
### `telefuser serve` — Batch Request-Response Mode
72+
73+
Use for batch text-to-video, image-to-video, image generation, and super-resolution.
74+
75+
- Task-based API under `/v1/tasks/*`
76+
- OpenAI-compatible routes under `/v1/images` and `/v1/videos`
77+
- Pipeline contracts for structured parameter exposure
78+
- Async scheduling with request isolation
79+
80+
### `telefuser stream-serve` — Continuous Streaming Mode
81+
82+
Use for real-time world models, interactive generation, speech-driven animation, and streaming media.
83+
84+
- Server-push WebRTC for progressive video output
85+
- Bidirectional WebRTC for interactive control loops (DataChannel + RTP)
86+
- Stateful sessions with continuous chunk generation
87+
88+
See the [Stream Server Guide](stream_server.md) for full streaming documentation.
89+
90+
---
91+
92+
## Supported Pipelines
93+
94+
### World Model and Real-Time Oriented
95+
96+
| Pipeline | Task | Notes |
97+
|----------|------|-------|
98+
| `LingBot-World-Fast` | Bidirectional world-model streaming | Interactive WebRTC control loop — see [Stream Server Guide](stream_server.md) |
99+
| `LiveAct` | S2V (speech-to-video) | Speech-driven talking head generation |
100+
| `FlashVSR` | VSR | Streaming video super-resolution |
101+
| `LongCat-Video` | T2V, I2V, VC | Long-form generation and continuation |
102+
103+
### Video Generation
104+
105+
| Pipeline | Task | Notes |
106+
|----------|------|-------|
107+
| `WanVideo` (Wan2.1 / Wan2.2) | T2V, I2V, FL2V | Main video generation family |
108+
| `HunyuanVideo` | T2V, I2V | Supported via service examples |
109+
| `LTX Video` | I2V + Audio | Unified audio-video generation |
110+
111+
### Image Generation
112+
113+
| Pipeline | Task | Notes |
114+
|----------|------|-------|
115+
| `Qwen-Image` | T2I, Edit | Image generation and editing |
116+
| `Z-Image` | T2I | Image generation |
117+
| `Flux2 Klein` | T2I | Image generation |
118+
119+
---
120+
60121
## CLI Usage
61122

62123
The TeleFuser CLI provides commands for starting API servers, validating pipelines, and scanning for security issues.

0 commit comments

Comments
 (0)