let5see/AI-Video
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

chore: Merge upstream/main with RunningHub 48G support and FAQ improvements

Browse Source
This commit is contained in:
empty
2026-01-06 17:48:02 +08:00
parent ba96a59822 e1e5aaf130
commit 8db3693e08
12 changed files with 1828 additions and 2138 deletions
Download Patch File Download Diff File Expand all files Collapse all files

25
README.md
View File

@@ -31,22 +31,17 @@ https://github.com/user-attachments/assets/a42e7457-fcc8-40da-83fc-784c45a8b95d
![Web UI界面](resources/webui.png)
## 📋 更新日志
## 📋 最近更新
### 2025-12-04
- 🎨 **新增「自定义素材」功能** - 支持用户上传自己的照片和视频:
- 📸 **智能图片分析**: AI 自动识别图片内容,生成精准描述
- 🎥 **视频理解能力**: 支持视频素材分析,理解视频中的场景和内容
- ✍️ **智能脚本生成**: 基于素材内容和用户意图,自动生成视频文案和旁白
- 🎬 **一键成片**: 将你的照片/视频素材,快速制作成带文案和配音的完整视频
- 💡 降低创作门槛,让没有素材的用户也能轻松制作专业视频
### 2025-11-18
- 优化 RunningHub 服务调用，支持并行处理，大幅提升生成速度
- 新增历史记录页面，可查看和管理所有生成过的视频
- 支持一次性创建多个视频任务，提高批量创作效率
- ✅ **2026-01-06**: 新增 RunningHub 48G 显存机器调用支持
- ✅ **2025-12-28**: 支持 RunningHub 并发限制可配置，优化 LLM 返回结构化数据的逻辑
- ✅ **2025-12-17**: 支持 ComfyUI API Key 配置，支持 Nano Banana 模型调用，API 接口支持模板自定义参数
- ✅ **2025-12-10**: 侧边栏内置 FAQ，锁定 edge-tts 版本修复 TTS 服务不稳定问题
- ✅ **2025-12-08**: 支持固定脚本多种分割方式(段落/行/句子)，优化模板选择交互逻辑支持直接预览选择
- ✅ **2025-12-06**: 修复视频生成 API 返回 URL 路径处理，支持跨平台兼容
- ✅ **2025-12-05**: 新增 Windows 整合包下载，优化图片与视频反推工作流
- ✅ **2025-12-04**: 新增「自定义素材」功能，支持用户上传自己的照片和视频，AI 智能分析生成脚本
- ✅ **2025-11-18**: 优化 RunningHub 服务调用支持并行处理，新增历史记录页面，支持批量创建视频任务
## ✨ 功能亮点

25
README_EN.md
View File

@@ -30,22 +30,17 @@ Just input a **topic**, and Pixelle-Video will automatically:
![Web UI Interface](resources/webui_en.png)
## 📋 Changelog
## 📋 Recent Updates
### 2025-12-04
- 🎨 **New "Custom Media" Feature** - Support uploading your own photos and videos:
- 📸 **Smart Image Analysis**: AI automatically recognizes image content and generates accurate descriptions
- 🎥 **Video Understanding**: Supports video asset analysis to understand scenes and content
- ✍️ **Intelligent Script Generation**: Auto-generates video scripts and narration based on asset content and user intent
- 🎬 **One-Click Production**: Transform your photos/videos into complete videos with scripts and voiceover
- 💡 Lower the creative barrier for users without existing media assets
### 2025-11-18
- Optimized RunningHub service calls with parallel processing for significantly faster speed
- Added history page to view and manage all generated videos
- Support creating multiple video tasks at once for efficient batch production
- ✅ **2026-01-06**: Added RunningHub 48G VRAM machine support
- ✅ **2025-12-28**: Configurable RunningHub concurrency limit, improved LLM structured data response handling
- ✅ **2025-12-17**: Added ComfyUI API Key configuration, Nano Banana model support, API template custom parameters
- ✅ **2025-12-10**: Built-in FAQ in sidebar, fixed edge-tts version to resolve TTS service instability
- ✅ **2025-12-08**: Support multiple script split modes (paragraph/line/sentence), improved template selection with direct preview
- ✅ **2025-12-06**: Fixed video generation API URL path handling with cross-platform compatibility
- ✅ **2025-12-05**: Added Windows all-in-one package download, optimized image and video analysis workflows
- ✅ **2025-12-04**: New "Custom Media" feature - upload your photos/videos with AI-powered analysis and script generation
- ✅ **2025-11-18**: Parallel processing for RunningHub, added history page, batch video task creation support
## ✨ Key Features

7
pixelle_video/config/manager.py
View File

@@ -131,6 +131,7 @@ class ConfigManager:
"comfyui_api_key": self.config.comfyui.comfyui_api_key,
"runninghub_api_key": self.config.comfyui.runninghub_api_key,
"runninghub_concurrent_limit": self.config.comfyui.runninghub_concurrent_limit,
"runninghub_instance_type": self.config.comfyui.runninghub_instance_type,
"tts": {
"default_workflow": self.config.comfyui.tts.default_workflow,
},
@@ -149,7 +150,8 @@ class ConfigManager:
comfyui_url: Optional[str] = None,
comfyui_api_key: Optional[str] = None,
runninghub_api_key: Optional[str] = None,
runninghub_concurrent_limit: Optional[int] = None
runninghub_concurrent_limit: Optional[int] = None,
runninghub_instance_type: Optional[str] = None
):
"""Set ComfyUI global configuration"""
updates = {}
@@ -161,6 +163,9 @@ class ConfigManager:
updates["runninghub_api_key"] = runninghub_api_key
if runninghub_concurrent_limit is not None:
updates["runninghub_concurrent_limit"] = runninghub_concurrent_limit
if runninghub_instance_type is not None:
# Empty string means disable (treat as None for storage)
updates["runninghub_instance_type"] = runninghub_instance_type if runninghub_instance_type else None
if updates:
self.update({"comfyui": updates})

1
pixelle_video/config/schema.py
View File

@@ -74,6 +74,7 @@ class ComfyUIConfig(BaseModel):
comfyui_api_key: Optional[str] = Field(default=None, description="ComfyUI API Key (optional)")
runninghub_api_key: Optional[str] = Field(default=None, description="RunningHub API Key (optional)")
runninghub_concurrent_limit: int = Field(default=1, ge=1, le=10, description="RunningHub concurrent execution limit (1-10)")
runninghub_instance_type: Optional[str] = Field(default=None, description="RunningHub instance type (optional, set to 'plus' for 48GB VRAM)")
tts: TTSSubConfig = Field(default_factory=TTSSubConfig, description="TTS-specific configuration")
image: ImageSubConfig = Field(default_factory=ImageSubConfig, description="Image-specific configuration")
video: VideoSubConfig = Field(default_factory=VideoSubConfig, description="Video-specific configuration")

4
pixelle_video/service.py
View File

@@ -120,6 +120,10 @@ class PixelleVideoCore:
kit_config["api_key"] = comfyui_config["comfyui_api_key"]
if comfyui_config.get("runninghub_api_key"):
kit_config["runninghub_api_key"] = comfyui_config["runninghub_api_key"]
# Only pass instance_type if it has a non-empty value
instance_type = comfyui_config.get("runninghub_instance_type")
if instance_type and instance_type.strip():
kit_config["runninghub_instance_type"] = instance_type
return kit_config

12
pixelle_video/services/comfy_base_service.py
View File

@@ -231,6 +231,7 @@ class ComfyBaseService:
self,
comfyui_url: Optional[str] = None,
runninghub_api_key: Optional[str] = None,
runninghub_instance_type: Optional[str] = None,
) -> Dict[str, Any]:
"""
Prepare ComfyKit configuration
@@ -238,6 +239,7 @@ class ComfyBaseService:
Args:
comfyui_url: ComfyUI URL (optional, overrides config)
runninghub_api_key: RunningHub API key (optional, overrides config)
runninghub_instance_type: RunningHub instance type (optional, overrides config)
Returns:
ComfyKit configuration dict
@@ -262,6 +264,16 @@ class ComfyBaseService:
if final_rh_key:
kit_config["runninghub_api_key"] = final_rh_key
# RunningHub instance type (priority: param > global config > env)
# Only pass if non-empty value
final_instance_type = (
runninghub_instance_type
or self.global_config.get("runninghub_instance_type")
or os.getenv("RUNNINGHUB_INSTANCE_TYPE")
)
if final_instance_type and final_instance_type.strip():
kit_config["runninghub_instance_type"] = final_instance_type
logger.debug(f"ComfyKit config: {kit_config}")
return kit_config

2
pyproject.toml
View File

@@ -24,7 +24,7 @@ dependencies = [
"fastapi>=0.115.0",
"uvicorn[standard]>=0.32.0",
"python-multipart>=0.0.12",
"comfykit>=0.1.11",
"comfykit>=0.1.12",
"beautifulsoup4>=4.14.2",
]

3781
uv.lock generated
View File

File diff suppressed because it is too large Load Diff

58
web/components/faq.py
View File

@@ -14,6 +14,7 @@
FAQ component for displaying frequently asked questions
"""
import re
from pathlib import Path
from typing import Optional
@@ -57,6 +58,50 @@ def load_faq_content(language: str) -> Optional[str]:
return None
def parse_faq_sections(content: str) -> list[tuple[str, str]]:
"""
Parse FAQ content into sections by ### headings
Args:
content: Raw markdown content
Returns:
List of (question, answer) tuples
"""
# Remove the first main heading (starts with #, not ###)
lines = content.split('\n')
if lines and lines[0].startswith('#') and not lines[0].startswith('##'):
content = '\n'.join(lines[1:])
# Split by ### headings (top-level questions)
# Pattern matches ### at start of line followed by question text
pattern = r'^###\s+(.+?)$'
sections = []
current_question = None
current_answer_lines = []
for line in content.split('\n'):
match = re.match(pattern, line)
if match:
# Save previous section if exists
if current_question is not None:
answer = '\n'.join(current_answer_lines).strip()
sections.append((current_question, answer))
# Start new section
current_question = match.group(1).strip()
current_answer_lines = []
else:
current_answer_lines.append(line)
# Save last section
if current_question is not None:
answer = '\n'.join(current_answer_lines).strip()
sections.append((current_question, answer))
return sections
def render_faq_sidebar():
"""
Render FAQ in the sidebar
@@ -77,14 +122,13 @@ def render_faq_sidebar():
if faq_content:
# Display FAQ in an expander, expanded by default
with st.expander(tr('faq.expand_to_view', fallback='FAQ'), expanded=True):
# Remove the first heading from FAQ content since we already show it above
lines = faq_content.split('\n')
# Skip the first line if it's a heading
if lines and lines[0].startswith('#'):
faq_content = '\n'.join(lines[1:])
# Parse FAQ into sections
sections = parse_faq_sections(faq_content)
# Display FAQ content
st.markdown(faq_content, unsafe_allow_html=True)
# Display each question in its own collapsible expander
for question, answer in sections:
with st.expander(question, expanded=False):
st.markdown(answer, unsafe_allow_html=True)
# Add a link to GitHub issues for more help
st.markdown(

43
web/components/settings.py
View File

@@ -180,15 +180,35 @@ def render_advanced_settings():
f"(https://www.runninghub{'.cn' if get_language() == 'zh_CN' else '.ai'}/?inviteCode=bozpdlbj)"
)
# RunningHub concurrent limit
runninghub_concurrent_limit = st.number_input(
tr("settings.comfyui.runninghub_concurrent_limit"),
min_value=1,
max_value=10,
value=comfyui_config.get("runninghub_concurrent_limit", 1),
help=tr("settings.comfyui.runninghub_concurrent_limit_help"),
key="runninghub_concurrent_limit_input"
)
# RunningHub concurrent limit and instance type (in one row)
limit_col, instance_col = st.columns(2)
with limit_col:
runninghub_concurrent_limit = st.number_input(
tr("settings.comfyui.runninghub_concurrent_limit"),
min_value=1,
max_value=10,
value=comfyui_config.get("runninghub_concurrent_limit", 1),
help=tr("settings.comfyui.runninghub_concurrent_limit_help"),
key="runninghub_concurrent_limit_input"
)
with instance_col:
# Check if instance type is "plus" (48G VRAM enabled)
current_instance_type = comfyui_config.get("runninghub_instance_type") or ""
is_plus_enabled = current_instance_type == "plus"
# Instance type options with i18n
instance_options = [
tr("settings.comfyui.runninghub_instance_24g"),
tr("settings.comfyui.runninghub_instance_48g"),
]
runninghub_instance_type_display = st.selectbox(
tr("settings.comfyui.runninghub_instance_type"),
options=instance_options,
index=1 if is_plus_enabled else 0,
help=tr("settings.comfyui.runninghub_instance_type_help"),
key="runninghub_instance_type_input"
)
# Convert display value back to actual value
runninghub_48g_enabled = runninghub_instance_type_display == tr("settings.comfyui.runninghub_instance_48g")
# ====================================================================
# Action Buttons (full width at bottom)
@@ -206,11 +226,14 @@ def render_advanced_settings():
config_manager.set_llm_config(llm_api_key, llm_base_url, llm_model)
# Save ComfyUI configuration (optional fields, always save what's provided)
# Convert checkbox to instance type: True -> "plus", False -> ""
instance_type = "plus" if runninghub_48g_enabled else ""
config_manager.set_comfyui_config(
comfyui_url=comfyui_url if comfyui_url else None,
comfyui_api_key=comfyui_api_key if comfyui_api_key else None,
runninghub_api_key=runninghub_api_key if runninghub_api_key else None,
runninghub_concurrent_limit=int(runninghub_concurrent_limit)
runninghub_concurrent_limit=int(runninghub_concurrent_limit),
runninghub_instance_type=instance_type
)
# Only save to file if LLM config is valid

4
web/i18n/locales/en_US.json
View File

@@ -197,6 +197,10 @@
"settings.comfyui.runninghub_get_api_key": "Get RunningHub API Key",
"settings.comfyui.runninghub_concurrent_limit": "Concurrent Limit",
"settings.comfyui.runninghub_concurrent_limit_help": "RunningHub concurrent execution limit (1-10), default is 1 for regular members, adjust based on your membership level",
"settings.comfyui.runninghub_instance_type": "Machine Spec",
"settings.comfyui.runninghub_instance_type_help": "Select RunningHub machine spec, 48G VRAM is suitable for large models or high-resolution generation (requires membership support)",
"settings.comfyui.runninghub_instance_24g": "24G VRAM",
"settings.comfyui.runninghub_instance_48g": "48G VRAM",
"tts.inference_mode": "Synthesis Mode",
"tts.mode.local": "Local Synthesis",
"tts.mode.comfyui": "ComfyUI Synthesis",

4
web/i18n/locales/zh_CN.json
View File

@@ -197,6 +197,10 @@
"settings.comfyui.runninghub_get_api_key": "点此获取 RunningHub API Key",
"settings.comfyui.runninghub_concurrent_limit": "并发限制",
"settings.comfyui.runninghub_concurrent_limit_help": "RunningHub 并发执行数量（1-10），普通会员默认为1，请根据您的会员等级调整",
"settings.comfyui.runninghub_instance_type": "机器规格",
"settings.comfyui.runninghub_instance_type_help": "选择 RunningHub 机器规格，48G 显存适用于大模型或高分辨率生成（需要会员支持）",
"settings.comfyui.runninghub_instance_24g": "24G 显存",
"settings.comfyui.runninghub_instance_48g": "48G 显存",
"tts.inference_mode": "合成方式",
"tts.mode.local": "本地合成",
"tts.mode.comfyui": "ComfyUI 合成",