AI-kfc/2025-02-13-19-00-11.png
... ...
Binary files /dev/null and b/AI-kfc/2025-02-13-19-00-11.png differ
AI-kfc/AI-audio/AI-audio.md
... ...
@@ -0,0 +1,6 @@
1
+
2
+# AI-audio
3
+
4
+pip install openai-whisper
5
+
6
+pip install git+https://github.com/openai/whisper.git
... ...
\ No newline at end of file
AI-kfc/AI-capture-screen/AI-capture-screen.md
... ...
@@ -0,0 +1,14 @@
1
+
2
+# AI-capture-screen
3
+
4
+A tool for localized automatic capture and analysis of screen/audio data and building personalized AI applications: Screenpipe, http://Rewind.ai open source alternative
5
+
6
+Features:
7
+
8
+1. 24/7 capture of screen and audio input/output, no manual operation required
9
+
10
+2. Build personalized AI, you can build AI applications based on what you see, say, and hear
11
+
12
+3. All data is saved locally, supporting integration with Ollama, etc.
13
+
14
+github: https://github.com/mediar-ai/screenpipe
... ...
\ No newline at end of file
AI-kfc/AI-code/cursor.md
... ...
@@ -0,0 +1,6 @@
1
+
2
+# cursor.md
3
+
4
+Awesome CursorRules (https://github.com/PatrickJS/awesome-cursorrules):
5
+
6
+A curated list of .cursorrules files to improve the Cursor AI code editing experience, support custom AI behavior, maintain code consistency, enhance project context awareness, improve productivity and team collaboration
... ...
\ No newline at end of file
AI-kfc/AI-image/AI-image.md
... ...
@@ -0,0 +1,21 @@
1
+
2
+# AI-image
3
+
4
+## consistent
5
+
6
+Hahahaha, I recommend a character consistency rendering platform: https://rendernet.ai
7
+I played with it for a while and the effect is pretty good. Lin Junjie looks so handsome riding a Harley motorcycle! 😎
8
+💡Massive style library can customize different styles
9
+💡Control the character's posture
10
+💡10 free generation opportunities per day
11
+💡2 images can be generated each time
12
+You can try it, but... listen to my advice, don't mess around! Okay? #AI
13
+
14
+
15
+## image promot
16
+
17
+🗣Vave BG: AI image library, each image will display the generated prompt words and AI tools, can be downloaded for free and used for commercial and personal purposes
18
+
19
+🏷Tags: #AI #Image #Commercially available #Material
20
+
21
+👉Link: https://www.appmiu.com/26673.html
... ...
\ No newline at end of file
AI-kfc/AI-video/Video_note_generator/Video_note_generator.md
... ...
@@ -0,0 +1,53 @@
1
+
2
+# Video_note_generator
3
+
4
+
5
+#小红书
6
+
7
+Use Python to convert videos into high-quality Xiaohongshu notes in one click, automatically optimize content and illustrations. The code is open source. Thanks to the author Xuanqing
8
+
9
+Application scenarios
10
+
11
+- Content creators: quickly convert videos/live content into articles
12
+- Knowledge managers: automatically organize video notes and learning points
13
+- Social media operations: batch generate high-quality Xiaohongshu notes
14
+
15
+https://github.com/whotto/Video_note_generator
16
+
17
+ git clone git@github.com:whotto/Video_note_generator.git
18
+
19
+ pip install -r requirements.txt
20
+
21
+ cp .env.example .env
22
+
23
+- [[openrouter]] - [[unsplash]]
24
+
25
+注意：以下环境变量未设置：
26
+ - UNSPLASH_REDIRECT_URI (用于Unsplash回调)
27
+
28
+python video_note_generator.py urls.txt
29
+
30
+
31
+
32
+## requirements
33
+
34
+
35
+pip install yt-dlp
36
+
37
+ openai>=1.0.0
38
+ httpx>=0.24.1
39
+ yt-dlp>=2023.11.16
40
+ openai-whisper>=2023.11.17
41
+ python-dotenv>=1.0.0
42
+ requests>=2.31.0
43
+ beautifulsoup4>=4.12.2
44
+ python-unsplash>=1.1.0
45
+ Pillow>=10.1.0
46
+ urllib3>=2.1.0
47
+ certifi>=2023.11.17
48
+ ffmpeg-python>=0.2.0
49
+
50
+ # Optional dependencies for better performance
51
+ torch>=2.1.0
52
+ pytube>=15.0.0
53
+ you-get>=0.4.1650
AI-kfc/AI-video/ai-video.md
... ...
@@ -0,0 +1,16 @@
1
+
2
+# ai-vid.md
3
+
4
+## AI-Youtube-Shorts-Generator
5
+
6
+Wow, you can slice the essence of a long video into a film with one click? This open source software is crazy:
7
+
8
+https://github.com/SamurAIGPT/AI-Youtube-Shorts-Generator
9
+
10
+The news of OpenAI's transformation to profitability was a hot topic a few days ago. A lot of people were discussing, what is their focus? As long as the most essential slice is cut out, how can it be automatically cut out?
11
+
12
+https://x.com/huangyun_122/status/1841431233261117786
13
+
14
+
15
+- [[Video_note_generator]]
16
+
AI-kfc/Nividia/2024-12-16-14-34-35.png
... ...
Binary files /dev/null and b/AI-kfc/Nividia/2024-12-16-14-34-35.png differ
AI-kfc/Nividia/Nvidia.md
... ...
@@ -0,0 +1,20 @@
1
+
2
+# Nvidia
3
+
4
+
5
+## Build - Stable Diffusion XL
6
+
7
+Stable Diffusion XL is the latest version of the painting model released by Stability AI. You only need to describe it in words, and you can let the computer help you draw. In general, you need to configure your own computing power (such as a high-performance graphics card), but NVIDIA provides an online, free Stable Diffusion XL service that is ready to use.
8
+
9
+This is a free instance hosted on NVIDIA NGC (NVIDIA GPU Cloud, a GPU-accelerated cloud platform optimized for deep learning and scientific computing).
10
+
11
+Just simply enter the Positive prompt and Negative prompt:
12
+
13
+![](2024-12-16-14-34-35.png)
14
+
15
+[build-stable-diffusion-xl](https://build.nvidia.com/stabilityai/stable-diffusion-xl)
16
+
17
+
18
+## NGC
19
+
20
+[NGC-nvidia](https://catalog.ngc.nvidia.com/?filters=&orderBy=weightPopularDESC&query=&page=&pageSize=)
... ...
\ No newline at end of file
AI-kfc/ai-blog/2024-12-30-03-56-16.png
... ...
Binary files /dev/null and b/AI-kfc/ai-blog/2024-12-30-03-56-16.png differ
AI-kfc/ai-blog/ai-blog.md
... ...
@@ -0,0 +1,14 @@
1
+
2
+# ai-blog
3
+
4
+Blog artifact! Voice-to-blog tool: Robo Blogger, it will automatically turn your words into blog posts, and automatically add reference materials if necessary
5
+
6
+This can be used to write in fragmented time, such as when walking or driving, you have a good idea, you can use the voice-to-text tool to record and input it, it can help you plan the article structure and produce professional articles
7
+
8
+Supports citing reference documents, supports different types of blog templates, and supports custom blog structure
9
+
10
+#AI blog #voice-to-blog #blog assistant #RoboBlogger
11
+
12
+https://github.com/langchain-ai/robo-blogger
13
+
14
+![](2024-12-30-03-56-16.png)
... ...
\ No newline at end of file
AI-kfc/ai-browser/ai-browser.md
... ...
@@ -0,0 +1,12 @@
1
+
2
+# ai-browser
3
+
4
+🛜 Browser Use
5
+
6
+Make websites accessible for AI agents 🤖
7
+
8
+Browser use is the easiest way to connect your AI agents with the browser
9
+
10
+Use any LLM supported by LangChain (e.g. gpt4o, gpt4o mini, claude 3.5 sonnet, llama 3.1 405b, etc.)
11
+
12
+https://github.com/gregpr07/browser-use
... ...
\ No newline at end of file
AI-kfc/ai-chatbot.md
... ...
@@ -0,0 +1,23 @@
1
+
2
+# ai-chatbot.md
3
+
4
+## chatwise
5
+
6
+[ChatWise: The second fastest AI chatbot](https://chatwise.app/)
7
+
8
+Supports multiple LLM models such as ChatGPT, DeepSeek, Claude, Gemini and Llama
9
+
10
+Supports web search and rendering HTML/React/charts/documents
11
+
12
+## cherry studio
13
+
14
+After trying various knowledge base building solutions (of course, I can do it), the most convenient one is [Cherry Studio](https://github.com/CherryHQ/cherry-studio), which only takes four steps:
15
+
16
+1⃣Download the nomic-embed-text model from Ollama
17
+2⃣Select the corresponding embedding model
18
+3⃣Upload the document
19
+4⃣Set the knowledge base in the dialogue and call DeepSeek-R1
20
+
21
+If your graphics card is not powerful, don't run the local model, just use the free quota, trust me.
22
+
23
+![](2025-02-13-19-00-11.png)
... ...
\ No newline at end of file
AI-kfc/ai-search/2025-02-13-18-57-19.png
... ...
Binary files /dev/null and b/AI-kfc/ai-search/2025-02-13-18-57-19.png differ
AI-kfc/ai-search/ai-search.md
... ...
@@ -0,0 +1,22 @@
1
+
2
+# ai-search
3
+
4
+- https://www.qoodle.ai/
5
+https://preview.redd.it/rg6gze1qz39e1.gif?width=1920&format=mp4&s=05f854f6b32dd460fbfc8013adb6bf7d019d654e
6
+
7
+- https://github.com/itsPreto/tangent
8
+
9
+
10
+
11
+
12
+## Xyne
13
+
14
+An intelligent search and Q&A tool for work: [Xyne](https://xynehq.com/), which integrates various data at work, can efficiently search for various files, emails, chat records, application tools and other data and reply to questions and answers
15
+
16
+It can be used to assist in finding files, checking customer/project information, finding problem solutions, etc.
17
+
18
+It can connect commonly used tools at work, all functions of Google Workspace, Slack, GitHub, etc., integrate the data of these applications together for easy search
19
+
20
+Support local and cloud deployment, support multiple models, support multi-threaded data ingestion
21
+
22
+![](2025-02-13-18-57-19.png)
... ...
\ No newline at end of file
AI-kfc/ai-trans/2024-12-30-03-39-32.png
... ...
Binary files /dev/null and b/AI-kfc/ai-trans/2024-12-30-03-39-32.png differ
AI-kfc/ai-trans/ai-trans.md
... ...
@@ -0,0 +1,31 @@
1
+
2
+# ai-trans
3
+
4
+## EasyVideoTrans
5
+
6
+What is EasyVideoTrans?
7
+
8
+EasyVideoTrans is a free, open source and easy-to-use AI video translation and dubbing tool that can help users quickly convert videos into Chinese and other languages. It is currently the fastest English video to Chinese tool, fully automatic one-stop operation, from video download, audio extraction, subtitle extraction, Chinese subtitle dubbing, voice connection and rendering preview video all one-click operation, Xiaobai also suitable for AI video translation and dubbing tool
9
+
10
+
11
+![](2024-12-30-03-39-32.png)
12
+
13
+How to use EasyVideoTrans
14
+
15
+Official website: https://easyvideotrans.com/
16
+
17
+Backend open source: https://github.com/sutro-planet/easyvideotrans-frontend
18
+
19
+Front-end open source: https://github.com/sutro-planet/easyvideotrans-frontend
20
+
21
+ptvzhen: https://github.com/CuSO4Gem/pytvzhen
22
+
23
+
24
+
25
+## pyvideotrans
26
+
27
+https://github.com/jianchang512/pyvideotrans
28
+
29
+About
30
+
31
+Translate the video from one language to another and add dubbing. 将视频从一种语言翻译为另一种语言，同时支持语音识别转录、语音合成、字幕翻译。
... ...
\ No newline at end of file
AI-kfc/ali/2025-03-02-20-10-37.png
... ...
Binary files /dev/null and b/AI-kfc/ali/2025-03-02-20-10-37.png differ
AI-kfc/ali/ali.md
... ...
@@ -0,0 +1,26 @@
1
+
2
+# ali
3
+
4
+
5
+## qwen2.5
6
+
7
+Qwen: Alibaba's Qwen 2.5-Max web version is better than DeepSeek V3 Supports deep thinking and online search, supports online model selection, etc.
8
+
9
+![](2025-03-02-20-10-37.png)
10
+
11
+
12
+https://chat.qwen.ai/
13
+
14
+## Qwen2.5-Coder-32B
15
+
16
+Alibaba has open-sourced the Qwen2.5-Coder-32B code model.
17
+
18
+The code capacity exceeds 40, close to Claude 3.5
19
+
20
+The entire series also has several specifications of 0.5B / 1.5B / 3B / 7B / 14B.
21
+
22
+In addition, quantization models in common formats are provided.
23
+
24
+There is also an open-source tool similar to Claude Artifacts, which you can play with.
25
+
26
+Details: https://qwenlm.github.io/blog/qwen2.5-coder-family/
... ...
\ No newline at end of file
AI-kfc/bytedance/bytedance.md
... ...
@@ -0,0 +1,10 @@
1
+
2
+# bytedance
3
+
4
+ByteDance's general image editing model SeedEdit has been released. It is called an editing model because it can modify the image based on the newly generated image, including photo editing, dressing, beautification, style conversion, adding and deleting elements in the specified area, and other editing operations.
5
+
6
+The biggest difference from ordinary editing is natural language communication. I tried it and it was quite smooth. Natural language-driven editing is still rare in the industry, especially considering that Doubao is one of the most influential AI general products at present. This time it may really affect the way many people design.
7
+
8
+I used Midjourney before. Like many friends, I urgently need a coherent and consistent drawing form that can meet the needs of long-term coherent design expressions, such as making comics with the same protagonist, different presentations of the same poster, etc.
9
+
10
+You can experience it on Doubao's official website doubao.com, you can try it.
... ...
\ No newline at end of file
AI-kfc/bytedance/douban/douban.md
... ...
@@ -0,0 +1,3 @@
1
+
2
+# douban
3
+
AI-kfc/digital-human/digital-human.md
... ...
@@ -0,0 +1,8 @@
1
+
2
+# digital-human
3
+
4
+🗣AWESOME-DIGITAL-HUMAN: Based on Dify's open source AI digital human tool, create a digital human with warmth and inject soul into the digital human
5
+
6
+🏷Tags: #sourcecode #AI #digitalhuman #self-deployment
7
+
8
+👉Link: https://www.appmiu.com/26593.html
... ...
\ No newline at end of file
AI-kfc/google/gemini/1899911888378749137_1.mp4
... ...
Binary files /dev/null and b/AI-kfc/google/gemini/1899911888378749137_1.mp4 differ
AI-kfc/google/gemini/1899996895776563620_1.mp4
... ...
Binary files /dev/null and b/AI-kfc/google/gemini/1899996895776563620_1.mp4 differ
AI-kfc/google/gemini/gemini.md
... ...
@@ -0,0 +1,20 @@
1
+
2
+# gemini
3
+
4
+### Gemini 2.0 Flash Native Image Generation
5
+![1899976207501807664_1](../../vps/ssh/1899976207501807664_1.jpg)
6
+Gemini 2.0 Flash now features native image generation, allowing users to create and edit images through chat. It supports text rendering for creating posters and can modify specific parts of an image without affecting others, making it easy to generate advertising posters with text and product images.
7
+
8
+![1899996895776563620_1](1899996895776563620_1.mp4)
9
+
10
+[ref](https://t.co/JECkZC7ZfR)
11
+
12
+
13
+### Gemini 2.0 Flash Image Generation
14
+
15
+AI Studio now offers Gemini 2.0 Flash exp, enabling the generation of natural images with consistent characters and the ability to create multiple images from a single prompt. It can generate images for story scenes, maintaining character consistency across images.
16
+
17
+![1899911888378749137_1](1899911888378749137_1.mp4)
18
+
19
+[ref](https://t.co/PJmraLkMCM)
20
+
AI-kfc/google/google-ai.md
... ...
@@ -0,0 +1,20 @@
1
+
2
+# google-ai.md
3
+
4
+Absolutely crazy! Google just fully released their Ai Lab: https://labs.google/fx/ .
5
+
6
+You can find
7
+
8
+- VidoFX
9
+- ImageFX
10
+- Music FX
11
+- Music FX DJ
12
+- Whisk
13
+
14
+Which basically means, using prompt, you can now generate:
15
+
16
+- Image
17
+- Music
18
+- Video
19
+
20
+All in one, in One Place!
... ...
\ No newline at end of file
AI/2025-02-13-19-00-11.png
... ...
Binary files a/AI/2025-02-13-19-00-11.png and /dev/null differ
AI/AI-audio/AI-audio.md
... ...
@@ -1,6 +0,0 @@
1
-
2
-# AI-audio
3
-
4
-pip install openai-whisper
5
-
6
-pip install git+https://github.com/openai/whisper.git
... ...
\ No newline at end of file
AI/AI-capture-screen/AI-capture-screen.md
... ...
@@ -1,14 +0,0 @@
1
-
2
-# AI-capture-screen
3
-
4
-A tool for localized automatic capture and analysis of screen/audio data and building personalized AI applications: Screenpipe, http://Rewind.ai open source alternative
5
-
6
-Features:
7
-
8
-1. 24/7 capture of screen and audio input/output, no manual operation required
9
-
10
-2. Build personalized AI, you can build AI applications based on what you see, say, and hear
11
-
12
-3. All data is saved locally, supporting integration with Ollama, etc.
13
-
14
-github: https://github.com/mediar-ai/screenpipe
... ...
\ No newline at end of file
AI/AI-code/cursor.md
... ...
@@ -1,6 +0,0 @@
1
-
2
-# cursor.md
3
-
4
-Awesome CursorRules (https://github.com/PatrickJS/awesome-cursorrules):
5
-
6
-A curated list of .cursorrules files to improve the Cursor AI code editing experience, support custom AI behavior, maintain code consistency, enhance project context awareness, improve productivity and team collaboration
... ...
\ No newline at end of file
AI/AI-image/AI-image.md
... ...
@@ -1,21 +0,0 @@
1
-
2
-# AI-image
3
-
4
-## consistent
5
-
6
-Hahahaha, I recommend a character consistency rendering platform: https://rendernet.ai
7
-I played with it for a while and the effect is pretty good. Lin Junjie looks so handsome riding a Harley motorcycle! 😎
8
-💡Massive style library can customize different styles
9
-💡Control the character's posture
10
-💡10 free generation opportunities per day
11
-💡2 images can be generated each time
12
-You can try it, but... listen to my advice, don't mess around! Okay? #AI
13
-
14
-
15
-## image promot
16
-
17
-🗣Vave BG: AI image library, each image will display the generated prompt words and AI tools, can be downloaded for free and used for commercial and personal purposes
18
-
19
-🏷Tags: #AI #Image #Commercially available #Material
20
-
21
-👉Link: https://www.appmiu.com/26673.html
... ...
\ No newline at end of file
AI/AI-video/Video_note_generator/Video_note_generator.md
... ...
@@ -1,53 +0,0 @@
1
-
2
-# Video_note_generator
3
-
4
-
5
-#小红书
6
-
7
-Use Python to convert videos into high-quality Xiaohongshu notes in one click, automatically optimize content and illustrations. The code is open source. Thanks to the author Xuanqing
8
-
9
-Application scenarios
10
-
11
-- Content creators: quickly convert videos/live content into articles
12
-- Knowledge managers: automatically organize video notes and learning points
13
-- Social media operations: batch generate high-quality Xiaohongshu notes
14
-
15
-https://github.com/whotto/Video_note_generator
16
-
17
- git clone git@github.com:whotto/Video_note_generator.git
18
-
19
- pip install -r requirements.txt
20
-
21
- cp .env.example .env
22
-
23
-- [[openrouter]] - [[unsplash]]
24
-
25
-注意：以下环境变量未设置：
26
- - UNSPLASH_REDIRECT_URI (用于Unsplash回调)
27
-
28
-python video_note_generator.py urls.txt
29
-
30
-
31
-
32
-## requirements
33
-
34
-
35
-pip install yt-dlp
36
-
37
- openai>=1.0.0
38
- httpx>=0.24.1
39
- yt-dlp>=2023.11.16
40
- openai-whisper>=2023.11.17
41
- python-dotenv>=1.0.0
42
- requests>=2.31.0
43
- beautifulsoup4>=4.12.2
44
- python-unsplash>=1.1.0
45
- Pillow>=10.1.0
46
- urllib3>=2.1.0
47
- certifi>=2023.11.17
48
- ffmpeg-python>=0.2.0
49
-
50
- # Optional dependencies for better performance
51
- torch>=2.1.0
52
- pytube>=15.0.0
53
- you-get>=0.4.1650
AI/AI-video/ai-video.md
... ...
@@ -1,16 +0,0 @@
1
-
2
-# ai-vid.md
3
-
4
-## AI-Youtube-Shorts-Generator
5
-
6
-Wow, you can slice the essence of a long video into a film with one click? This open source software is crazy:
7
-
8
-https://github.com/SamurAIGPT/AI-Youtube-Shorts-Generator
9
-
10
-The news of OpenAI's transformation to profitability was a hot topic a few days ago. A lot of people were discussing, what is their focus? As long as the most essential slice is cut out, how can it be automatically cut out?
11
-
12
-https://x.com/huangyun_122/status/1841431233261117786
13
-
14
-
15
-- [[Video_note_generator]]
16
-
AI/Nividia/2024-12-16-14-34-35.png
... ...
Binary files a/AI/Nividia/2024-12-16-14-34-35.png and /dev/null differ
AI/Nividia/Nvidia.md
... ...
@@ -1,20 +0,0 @@
1
-
2
-# Nvidia
3
-
4
-
5
-## Build - Stable Diffusion XL
6
-
7
-Stable Diffusion XL is the latest version of the painting model released by Stability AI. You only need to describe it in words, and you can let the computer help you draw. In general, you need to configure your own computing power (such as a high-performance graphics card), but NVIDIA provides an online, free Stable Diffusion XL service that is ready to use.
8
-
9
-This is a free instance hosted on NVIDIA NGC (NVIDIA GPU Cloud, a GPU-accelerated cloud platform optimized for deep learning and scientific computing).
10
-
11
-Just simply enter the Positive prompt and Negative prompt:
12
-
13
-![](2024-12-16-14-34-35.png)
14
-
15
-[build-stable-diffusion-xl](https://build.nvidia.com/stabilityai/stable-diffusion-xl)
16
-
17
-
18
-## NGC
19
-
20
-[NGC-nvidia](https://catalog.ngc.nvidia.com/?filters=&orderBy=weightPopularDESC&query=&page=&pageSize=)
... ...
\ No newline at end of file
AI/ai-blog/2024-12-30-03-56-16.png
... ...
Binary files a/AI/ai-blog/2024-12-30-03-56-16.png and /dev/null differ
AI/ai-blog/ai-blog.md
... ...
@@ -1,14 +0,0 @@
1
-
2
-# ai-blog
3
-
4
-Blog artifact! Voice-to-blog tool: Robo Blogger, it will automatically turn your words into blog posts, and automatically add reference materials if necessary
5
-
6
-This can be used to write in fragmented time, such as when walking or driving, you have a good idea, you can use the voice-to-text tool to record and input it, it can help you plan the article structure and produce professional articles
7
-
8
-Supports citing reference documents, supports different types of blog templates, and supports custom blog structure
9
-
10
-#AI blog #voice-to-blog #blog assistant #RoboBlogger
11
-
12
-https://github.com/langchain-ai/robo-blogger
13
-
14
-![](2024-12-30-03-56-16.png)
... ...
\ No newline at end of file
AI/ai-browser/ai-browser.md
... ...
@@ -1,12 +0,0 @@
1
-
2
-# ai-browser
3
-
4
-🛜 Browser Use
5
-
6
-Make websites accessible for AI agents 🤖
7
-
8
-Browser use is the easiest way to connect your AI agents with the browser
9
-
10
-Use any LLM supported by LangChain (e.g. gpt4o, gpt4o mini, claude 3.5 sonnet, llama 3.1 405b, etc.)
11
-
12
-https://github.com/gregpr07/browser-use
... ...
\ No newline at end of file
AI/ai-chatbot.md
... ...
@@ -1,23 +0,0 @@
1
-
2
-# ai-chatbot.md
3
-
4
-## chatwise
5
-
6
-[ChatWise: The second fastest AI chatbot](https://chatwise.app/)
7
-
8
-Supports multiple LLM models such as ChatGPT, DeepSeek, Claude, Gemini and Llama
9
-
10
-Supports web search and rendering HTML/React/charts/documents
11
-
12
-## cherry studio
13
-
14
-After trying various knowledge base building solutions (of course, I can do it), the most convenient one is [Cherry Studio](https://github.com/CherryHQ/cherry-studio), which only takes four steps:
15
-
16
-1⃣Download the nomic-embed-text model from Ollama
17
-2⃣Select the corresponding embedding model
18
-3⃣Upload the document
19
-4⃣Set the knowledge base in the dialogue and call DeepSeek-R1
20
-
21
-If your graphics card is not powerful, don't run the local model, just use the free quota, trust me.
22
-
23
-![](2025-02-13-19-00-11.png)
... ...
\ No newline at end of file
AI/ai-search/2025-02-13-18-57-19.png
... ...
Binary files a/AI/ai-search/2025-02-13-18-57-19.png and /dev/null differ
AI/ai-search/ai-search.md
... ...
@@ -1,22 +0,0 @@
1
-
2
-# ai-search
3
-
4
-- https://www.qoodle.ai/
5
-https://preview.redd.it/rg6gze1qz39e1.gif?width=1920&format=mp4&s=05f854f6b32dd460fbfc8013adb6bf7d019d654e
6
-
7
-- https://github.com/itsPreto/tangent
8
-
9
-
10
-
11
-
12
-## Xyne
13
-
14
-An intelligent search and Q&A tool for work: [Xyne](https://xynehq.com/), which integrates various data at work, can efficiently search for various files, emails, chat records, application tools and other data and reply to questions and answers
15
-
16
-It can be used to assist in finding files, checking customer/project information, finding problem solutions, etc.
17
-
18
-It can connect commonly used tools at work, all functions of Google Workspace, Slack, GitHub, etc., integrate the data of these applications together for easy search
19
-
20
-Support local and cloud deployment, support multiple models, support multi-threaded data ingestion
21
-
22
-![](2025-02-13-18-57-19.png)
... ...
\ No newline at end of file
AI/ai-trans/2024-12-30-03-39-32.png
... ...
Binary files a/AI/ai-trans/2024-12-30-03-39-32.png and /dev/null differ
AI/ai-trans/ai-trans.md
... ...
@@ -1,31 +0,0 @@
1
-
2
-# ai-trans
3
-
4
-## EasyVideoTrans
5
-
6
-What is EasyVideoTrans?
7
-
8
-EasyVideoTrans is a free, open source and easy-to-use AI video translation and dubbing tool that can help users quickly convert videos into Chinese and other languages. It is currently the fastest English video to Chinese tool, fully automatic one-stop operation, from video download, audio extraction, subtitle extraction, Chinese subtitle dubbing, voice connection and rendering preview video all one-click operation, Xiaobai also suitable for AI video translation and dubbing tool
9
-
10
-
11
-![](2024-12-30-03-39-32.png)
12
-
13
-How to use EasyVideoTrans
14
-
15
-Official website: https://easyvideotrans.com/
16
-
17
-Backend open source: https://github.com/sutro-planet/easyvideotrans-frontend
18
-
19
-Front-end open source: https://github.com/sutro-planet/easyvideotrans-frontend
20
-
21
-ptvzhen: https://github.com/CuSO4Gem/pytvzhen
22
-
23
-
24
-
25
-## pyvideotrans
26
-
27
-https://github.com/jianchang512/pyvideotrans
28
-
29
-About
30
-
31
-Translate the video from one language to another and add dubbing. 将视频从一种语言翻译为另一种语言，同时支持语音识别转录、语音合成、字幕翻译。
... ...
\ No newline at end of file
AI/ali/2025-03-02-20-10-37.png
... ...
Binary files a/AI/ali/2025-03-02-20-10-37.png and /dev/null differ
AI/ali/ali.md
... ...
@@ -1,26 +0,0 @@
1
-
2
-# ali
3
-
4
-
5
-## qwen2.5
6
-
7
-Qwen: Alibaba's Qwen 2.5-Max web version is better than DeepSeek V3 Supports deep thinking and online search, supports online model selection, etc.
8
-
9
-![](2025-03-02-20-10-37.png)
10
-
11
-
12
-https://chat.qwen.ai/
13
-
14
-## Qwen2.5-Coder-32B
15
-
16
-Alibaba has open-sourced the Qwen2.5-Coder-32B code model.
17
-
18
-The code capacity exceeds 40, close to Claude 3.5
19
-
20
-The entire series also has several specifications of 0.5B / 1.5B / 3B / 7B / 14B.
21
-
22
-In addition, quantization models in common formats are provided.
23
-
24
-There is also an open-source tool similar to Claude Artifacts, which you can play with.
25
-
26
-Details: https://qwenlm.github.io/blog/qwen2.5-coder-family/
... ...
\ No newline at end of file
AI/bytedance/bytedance.md
... ...
@@ -1,10 +0,0 @@
1
-
2
-# bytedance
3
-
4
-ByteDance's general image editing model SeedEdit has been released. It is called an editing model because it can modify the image based on the newly generated image, including photo editing, dressing, beautification, style conversion, adding and deleting elements in the specified area, and other editing operations.
5
-
6
-The biggest difference from ordinary editing is natural language communication. I tried it and it was quite smooth. Natural language-driven editing is still rare in the industry, especially considering that Doubao is one of the most influential AI general products at present. This time it may really affect the way many people design.
7
-
8
-I used Midjourney before. Like many friends, I urgently need a coherent and consistent drawing form that can meet the needs of long-term coherent design expressions, such as making comics with the same protagonist, different presentations of the same poster, etc.
9
-
10
-You can experience it on Doubao's official website doubao.com, you can try it.
... ...
\ No newline at end of file
AI/bytedance/douban/douban.md
... ...
@@ -1,3 +0,0 @@
1
-
2
-# douban
3
-
AI/digital-human/digital-human.md
... ...
@@ -1,8 +0,0 @@
1
-
2
-# digital-human
3
-
4
-🗣AWESOME-DIGITAL-HUMAN: Based on Dify's open source AI digital human tool, create a digital human with warmth and inject soul into the digital human
5
-
6
-🏷Tags: #sourcecode #AI #digitalhuman #self-deployment
7
-
8
-👉Link: https://www.appmiu.com/26593.html
... ...
\ No newline at end of file
AI/google/gemini/1899911888378749137_1.mp4
... ...
Binary files a/AI/google/gemini/1899911888378749137_1.mp4 and /dev/null differ
AI/google/gemini/1899996895776563620_1.mp4
... ...
Binary files a/AI/google/gemini/1899996895776563620_1.mp4 and /dev/null differ
AI/google/gemini/gemini.md
... ...
@@ -1,20 +0,0 @@
1
-
2
-# gemini
3
-
4
-### Gemini 2.0 Flash Native Image Generation
5
-![1899976207501807664_1](../../vps/ssh/1899976207501807664_1.jpg)
6
-Gemini 2.0 Flash now features native image generation, allowing users to create and edit images through chat. It supports text rendering for creating posters and can modify specific parts of an image without affecting others, making it easy to generate advertising posters with text and product images.
7
-
8
-![1899996895776563620_1](1899996895776563620_1.mp4)
9
-
10
-[ref](https://t.co/JECkZC7ZfR)
11
-
12
-
13
-### Gemini 2.0 Flash Image Generation
14
-
15
-AI Studio now offers Gemini 2.0 Flash exp, enabling the generation of natural images with consistent characters and the ability to create multiple images from a single prompt. It can generate images for story scenes, maintaining character consistency across images.
16
-
17
-![1899911888378749137_1](1899911888378749137_1.mp4)
18
-
19
-[ref](https://t.co/PJmraLkMCM)
20
-
AI/google/google-ai.md
... ...
@@ -1,20 +0,0 @@
1
-
2
-# google-ai.md
3
-
4
-Absolutely crazy! Google just fully released their Ai Lab: https://labs.google/fx/ .
5
-
6
-You can find
7
-
8
-- VidoFX
9
-- ImageFX
10
-- Music FX
11
-- Music FX DJ
12
-- Whisk
13
-
14
-Which basically means, using prompt, you can now generate:
15
-
16
-- Image
17
-- Music
18
-- Video
19
-
20
-All in one, in One Place!
... ...
\ No newline at end of file
android-kfc/adb-kfc/adb-kfc.md
... ...
@@ -0,0 +1,7 @@
1
+
2
+# adb-kfc
3
+
4
+adb logcat > crashlog.txt
5
+
6
+
7
+- [[scrcpy]]
... ...
\ No newline at end of file
android-kfc/shizuka-kfc/2025-11-06-12-37-13.png
... ...
Binary files /dev/null and b/android-kfc/shizuka-kfc/2025-11-06-12-37-13.png differ
android-kfc/shizuka-kfc/2025-11-06-12-37-29.png
... ...
Binary files /dev/null and b/android-kfc/shizuka-kfc/2025-11-06-12-37-29.png differ
android-kfc/shizuka-kfc/pixel-ims/pixel-ims.md
... ...
@@ -0,0 +1,18 @@
1
+
2
+# pixel-ims
3
+
4
+https://github.com/kyujin-cho/pixel-volte-patch/releases
5
+
6
+- [[tiktok]]
7
+
8
+
9
+
10
+**QPR2 Beta 3** introduced a new limit making non-system apps unable to call overrideConfig with persistent=true. The method still works by setting persistent to false but configurations will be reset on every reboot. Previously, with persistent=true they are only reset after a system update.
11
+
12
+
13
+
14
+
15
+## ref
16
+
17
+- [[shizuka]]
18
+
android-kfc/shizuka-kfc/shizuka.md
... ...
@@ -0,0 +1,71 @@
1
+
2
+# shizuka
3
+
4
+
5
+
6
+
7
+
8
+
9
+https://www.v2ex.com/t/1166050
10
+
11
+- [[tiktok]]
12
+
13
+- [[pixel-ims]]
14
+
15
+- [[adb-kfc]]
16
+
17
+
18
+## app list
19
+
20
+https://github.com/timschneeb/awesome-shizuku
21
+
22
+
23
+
24
+## ims 2
25
+
26
+https://github.com/vvb2060/Ims/releases
27
+
28
+https://github.com/Turbo1123/TurboIMS
29
+
30
+https://github.com/tdrkDev/pixel-volte-patch
31
+
32
+
33
+## nrfr
34
+
35
+
36
+https://github.com/Ackites/Nrfr
37
+
38
+
39
+Nrfr+Shizuku 即可完美修改 Sim 卡国家码，解除 Tiktok 限制，亲测 100%有效。
40
+
41
+![](2025-11-06-12-37-13.png)
42
+
43
+Shizuku： 提供必要的权限支持（ adb 连接）
44
+
45
+Nrfr：修改 SIM 卡国家码为支持的地区（如 JP 、US 等）
46
+
47
+这 2 个软件最重要的还是启动 Shiziku ，如果是第一次使用 adb 连接，可能会遇到一些问题。
48
+
49
+实测教程（有视频）：
50
+
51
+https://zhuanlan.zhihu.com/p/1961885244765507996
52
+
53
+Nrfr 一键修改十分简单没什么好说的。
54
+
55
+启动 Shizuku 的 3 个必要步骤：
56
+
57
+1 、打开「开发者模式」，并打开「无线调试」
58
+
59
+2 、如果是小米、红米手机还需要设置通知样式为“原生样式”才能正常输入 [配对码] ，在通知栏上输入配对码，成功配对才能启动 Shizuku 。
60
+
61
+3 、在 Shizuku 中给 Nrfr 授权
62
+
63
+成功修改 Sim 卡信息之后，即可完美运行谷歌商店下载的 Tiktok 版本：
64
+
65
+
66
+![](2025-11-06-12-37-29.png)
67
+
68
+
69
+关于 Tiktok 、谷歌商店、Shizuku 、Adb 相关的问题，有问必答
70
+
71
+
shizuka/2025-11-06-12-37-13.png
... ...
Binary files a/shizuka/2025-11-06-12-37-13.png and /dev/null differ
shizuka/2025-11-06-12-37-29.png
... ...
Binary files a/shizuka/2025-11-06-12-37-29.png and /dev/null differ
shizuka/pixel-ims/pixel-ims.md
... ...
@@ -1,9 +0,0 @@
1
-
2
-# pixel-ims
3
-
4
-https://github.com/kyujin-cho/pixel-volte-patch/releases
5
-
6
-
7
-## tiktok modifiy
8
-
9
-https://github.com/Ackites/Nrfr/issues/61
... ...
\ No newline at end of file
shizuka/shizuka.md
... ...
@@ -1,52 +0,0 @@
1
-
2
-# shizuka
3
-
4
-
5
-https://www.v2ex.com/t/1166050
6
-
7
-- [[tiktok]]
8
-
9
-- [[pixel-ims]]
10
-
11
-
12
-https://github.com/timschneeb/awesome-shizuku
13
-
14
-
15
-
16
-
17
-https://github.com/Ackites/Nrfr
18
-
19
-
20
-Nrfr+Shizuku 即可完美修改 Sim 卡国家码，解除 Tiktok 限制，亲测 100%有效。
21
-
22
-![](2025-11-06-12-37-13.png)
23
-
24
-Shizuku： 提供必要的权限支持（ adb 连接）
25
-
26
-Nrfr：修改 SIM 卡国家码为支持的地区（如 JP 、US 等）
27
-
28
-这 2 个软件最重要的还是启动 Shiziku ，如果是第一次使用 adb 连接，可能会遇到一些问题。
29
-
30
-实测教程（有视频）：
31
-
32
-https://zhuanlan.zhihu.com/p/1961885244765507996
33
-
34
-Nrfr 一键修改十分简单没什么好说的。
35
-
36
-启动 Shizuku 的 3 个必要步骤：
37
-
38
-1 、打开「开发者模式」，并打开「无线调试」
39
-
40
-2 、如果是小米、红米手机还需要设置通知样式为“原生样式”才能正常输入 [配对码] ，在通知栏上输入配对码，成功配对才能启动 Shizuku 。
41
-
42
-3 、在 Shizuku 中给 Nrfr 授权
43
-
44
-成功修改 Sim 卡信息之后，即可完美运行谷歌商店下载的 Tiktok 版本：
45
-
46
-
47
-![](2025-11-06-12-37-29.png)
48
-
49
-
50
-关于 Tiktok 、谷歌商店、Shizuku 、Adb 相关的问题，有问必答
51
-
52
-
social-media/tiktok/tiktok.md
... ...
@@ -5,8 +5,41 @@
5 5
6 6
- [[shizuka]]
7 7
8
+
9
+## tiktok iphone sim-free
8 10
我美区 app store 下载的，插卡也可以播放啊。
9 11
10 12
@zl1995 #12 试了下小火箭可以，Surge 却不行。。难不成是规则问题？😅
11 13
12
-@lixiaobai913 iOS 使用小火箭能看是应为 iOS 系统不允许 App 去读取 Sim 卡信息，Android 系统会检测国内的手机卡的
... ...
\ No newline at end of file
0
+@lixiaobai913 iOS 使用小火箭能看是应为 iOS 系统不允许 App 去读取 Sim 卡信息，Android 系统会检测国内的手机卡的
1
+
2
+
3
+
4
+
5
+## tiktok android sim-free
6
+
7
+- [[pixel-ims]]
8
+
9
+
10
+https://github.com/Ackites/Nrfr/issues/61
11
+
12
+
13
+- uninstall [[tiktok-mod]]
14
+
15
+- install normal version tiktok
16
+
17
+- in pixel-ims expert mode, set
18
+
19
+ sim_country_iso_override_string == us
20
+ carrier_name_override_bool == true
21
+ carrier_name_string == T-Mobile
22
+
23
+- switch ON/OFF airplane mode to active it
24
+
25
+
26
+
27
+## ref
28
+
29
+- [[encrypt]]
30
+
31
+- [[social-media]]
... ...
\ No newline at end of file
vscode-kfc.md
... ...
@@ -0,0 +1,7 @@
1
+
2
+
3
+# vscode-kfc.md
4
+
5
+- [[vscode]]
6
+
7
+- [[encrypt]]
... ...
\ No newline at end of file