empty
92183b083b
feat: Add Qwen VL support for character analysis, configurable via VLM_PROVIDER
2026-01-07 09:29:43 +08:00
empty
be216eacad
fix: Increase VLM max_tokens to 2000 to avoid response truncation
2026-01-07 03:37:55 +08:00
empty
8d82cf91d5
fix: Auto-detect and use GLM-4V vision model for character analysis
2026-01-07 03:33:56 +08:00
empty
8c35b0066f
fix: Enhance VLM response parsing to handle markdown code blocks
2026-01-07 03:31:42 +08:00
empty
b3cf9e64e5
feat: Implement Character Memory V1 - VLM analysis and prompt injection
2026-01-07 03:08:29 +08:00
empty
2be9256c48
fix: Use OpenAI multimodal message format for VLM style extraction
2026-01-07 00:22:33 +08:00
empty
297f3ccda4
feat: Enhance StyleGuard with VLM-based style extraction for specific style_prefix
2026-01-07 00:16:57 +08:00
empty
2978622f7f
feat(P0): Enhance image prompt generation with core imagery extraction for better text-image alignment
2026-01-06 23:04:20 +08:00
empty
8db3693e08
chore: Merge upstream/main with RunningHub 48G support and FAQ improvements
2026-01-06 17:48:02 +08:00
puke
d6b2f02cce
新增runninghub的48G显存机器调用支持
2026-01-06 11:11:11 +08:00
empty
79a6c2ef3e
feat: Add inpainting (局部重绘) feature for timeline editor
...
- Add canvas-based mask drawing tools (brush, eraser, rect, lasso)
- Add undo/redo history support for mask editing
- Integrate inpainting UI into preview player
- Add backend API endpoint for inpainting requests
- Add MediaService.inpaint method with ComfyUI workflow support
- Add Flux inpainting workflows for selfhost and RunningHub
🤖 Generated with [Claude Code](https://claude.com/claude-code )
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com >
2026-01-05 23:44:51 +08:00
empty
56db9bf9d2
feat: Add hybrid quality evaluation system with CLIP and VLM support
...
- Add FeatureExtractor for CLIP-based image/text feature extraction
- Add ObjectiveMetricsCalculator for technical quality metrics
- Add VLMEvaluator for vision language model evaluation
- Add HybridQualityGate combining objective + VLM evaluation
- Enhance CharacterMemory with visual feature support
- Add quality optional dependency (torch, ftfy, regex)
- Add unit tests for new modules
🤖 Generated with [Claude Code](https://claude.com/claude-code )
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com >
2026-01-05 15:56:44 +08:00
empty
ca018a9b1f
feat: Add comprehensive timeline editor with frame editing and regeneration capabilities
2026-01-05 14:48:43 +08:00
puke
c8f050eac8
优化FastAPI场景的runninghub并发设置问题
2025-12-28 16:34:01 +08:00
puke
c449a6dfe8
优化LLM返回结构化数据的逻辑
2025-12-28 15:06:34 +08:00
puke
1530d1f8c8
支持runninghub并发限制可配置
2025-12-28 14:56:22 +08:00
puke
cd09e8034f
优化llm配置的保存逻辑
2025-12-22 14:53:12 +08:00
puke
5b7ec15a9a
fix: https://github.com/AIDC-AI/Pixelle-Video/issues/35
2025-12-22 11:14:36 +08:00
puke
3e2a0976df
优化ComfyUI的API密钥保存逻辑
2025-12-17 19:03:55 +08:00
puke
c04b415ebd
支持ComfyUI ApiKey配置
2025-12-17 18:56:22 +08:00
puke
3cf6628022
支持固定脚本多种分割方式(段落/行/句子)，修复Edge TTS和模板切换问题
2025-12-08 16:59:02 +08:00
张欣
1769b7faee
Update prompts and content generators
2025-12-05 16:42:17 +08:00
strict235
3e1870581b
优化视频下载的文件命名逻辑
2025-12-04 23:54:09 +08:00
puke
584374aecb
优化语言一致性逻辑
2025-12-04 22:25:41 +08:00
puke
007a39c03a
开发基于视频素材生成视频的webui功能
2025-12-04 22:25:41 +08:00
puke
5c52696e6f
支持视频理解功能
2025-12-04 22:25:41 +08:00
puke
7425b9d23d
开发基于图片素材生成视频的webui功能
2025-12-04 22:25:41 +08:00
puke
ea784e0d06
支持基于图片素材的视频生成逻辑
2025-12-04 22:25:41 +08:00
puke
6e99612a68
重构pipeline的UI架构,支持后续pipeline的动态拓展
2025-12-04 22:25:41 +08:00
puke
3fd4696137
修复模板索引值传递问题
2025-11-27 17:13:22 +08:00
puke
5423b77ae8
对工作流和模板支持索引值参数输入
2025-11-27 11:40:40 +08:00
puke
a9e12d539b
确保临时、数据和输出目录在返回路径之前存在
2025-11-21 11:22:11 +08:00
puke
02ef878e3b
优化视频音频合并逻辑，新增智能时长调整功能
2025-11-21 00:56:24 +08:00
puke
d8e380bdb5
对分镜视频工作流传递duration参数
2025-11-21 00:32:22 +08:00
puke
7f904f6b19
修复视频尺寸传参未生效的问题
2025-11-20 20:09:53 +08:00
puke
d3f4a675db
统一RUNNING_HUB_PARALLEL_LIMIT控制逻辑
2025-11-20 14:42:59 +08:00
puke
fc70256152
默认禁用runninghub的并发(适配绝大多数的普通rh会员场景)
2025-11-20 00:55:32 +08:00
puke
8aad91006d
优化ComfyKit实例管理逻辑, 提升调用runninghub时的稳定性
2025-11-19 17:14:34 +08:00
puke
47baf1bf1e
修复视频参数设置失效问题
2025-11-19 15:09:18 +08:00
puke
dfa2f76c5b
支持历史生成结果的呈现逻辑
2025-11-18 17:51:56 +08:00
puke
fa8051d826
增加视频生成过程详细信息的持久化l; 优化session管理逻辑;
2025-11-17 17:56:15 +08:00
puke
c8df34a318
支持runninghub并行执行,提高视频生成效率
2025-11-17 14:33:51 +08:00
puke
d0c2d60cc5
添加资源下载的超时设置
2025-11-12 21:46:20 +08:00
puke
8a0e0dedd8
优化config中video的读取逻辑
2025-11-12 21:35:46 +08:00
puke
9eb8e7f82d
Merge branch 'dev_video'
2025-11-12 21:22:08 +08:00
puke
4785d1ddb5
优化本地调用edge-tts的证书逻辑
2025-11-12 20:57:29 +08:00
puke
f7e3162a4a
webui适配视频功能; 统一模板命名规范;
2025-11-12 20:01:09 +08:00
puke
0be64009a6
更改Qwen模型的apikey获取方式为百炼
2025-11-12 17:20:30 +08:00
puke
7443cbf9c2
模板中媒体尺寸改为预置方案
2025-11-12 17:19:06 +08:00
puke
64dcca204e
Merge branch 'main' into dev_video
2025-11-12 10:17:09 +08:00
