LangDrive是什么？

LangDrive 支持邮箱 & 客户服务联系 & 退款联系等

以下是 LangDrive 支持邮箱含客户服务: info@addy-ai.com .
LangDrive 公司信息

LangDrive 公司名字: Addy AI .
LangDrive Linkedin

LangDrive Linkedin链接: https://www.linkedin.com/company/addy-ai/
LangDrive Twitter

LangDrive Twitter链接: https://twitter.com/addyai_
LangDrive Github

LangDrive Github链接: https://github.com/addy-ai/langdrive

LangDrive 评价 (0)

5 满分 5 分

LangDrive数据分析

LangDrive 网站流量分析

社交媒体聆听

All

YouTube

Tiktok

搜索历史

14:25

How I Made My Own ChatGPT in 15 Minutes *IT WORKS!

In this video, I show you how to create your own ChatGPT without any knowledge of Artificial Intelligence (AI) or Machine Learning (ML) in less than 15 minutes. -------------------------------------------------------------------------------------------------- Code: https://github.com/addy-ai/langdrive/tree/main/demos/langdrive-chatgpt-demo -------------------------------------------------------------------------------------------------- Tech Stack - LangDrive (https://github.com/addy-ai/langdrive): Free and Open source library to train and use open source Large Language Models. I used LangDrive API to call a large language model called mistral-7b (https://huggingface.co/mistralai/Mistral-7B-v0.1). The Mistral-7B-v0.1 Large Language Model (LLM) is a pretrained generative text model with 7 billion parameters. Mistral-7B-v0.1 outperforms Llama 2 13B on all benchmarks we tested. - Screenshot to Code (https://github.com/abi/screenshot-to-code): Upload a screenshot of any website and get the frontend code for it. Used to create a lookalike of the ChatGPT webpage. - ChatGPT (https://chat.openai.com): Used to generate code to add interactivity for the webpage. FIND ME HERE -------------------------------------------------------------------------------------------------------- Twitter: https://twitter.com/michael_vandi Github: https://github.com/MichaelVandi/ Website: https://mvandi.com LinkedIn: https://www.linkedin.com/in/michael-vandi/ Instagram: https://www.instagram.com/mike_ehl_/ Discord: https://discord.gg/7dDeReJ -------------------------------------------------------------------------------------------------- ChatGPT Prompt Used "Act as a worldclass javascript developer. I am developing an AI chat application where a user asks a question, the content of the question is sent over to an AI model via an API, and the response from the AI is shown back to the user together with their question. I already have the frontend code for the webpage and I need you to use plain javascript to add interactivity to the frontend webpage. I will provide you with the front end code and your task is to finish the following tasks. 1. When a user click on the send button with id `send-button` Get the current value of the input element with id `message-input`. 2. Create the chat message div of class `chat-message` with it's child elements as currently specified in the fron end code. Give this chat message div a unique ID. For content of the child element of class "user-prompt" must be the value of the user's input found in the input element. For the AI response, just put a placeholder of "thinking" in the child element of class `ai-response`. 3. Append the chat message to the main content div of id `chat-history` 4. Make a POST request using the fetch API in javascript to the following model API endpoint and get the response: Here is the format of the model API. Model API format: ``` Endpoint: https://api.langdrive.ai/v1/chat/completions Request Body { "prompt": "string", "model": "string" "max_tokens": "number" } Response { "success": "boolean", "generated_text": "string" } ``` 5. Once the response is received from the model API endpoint, replace the "thinking" placeholder in the `ai-response` element in the current chat message element with the value of the response. 6. If there is an error or no response from the model. Create an error element saying `Sorry something went wrong. Please try again and append it in the `ai-response` element. Front end code ``` REPLACE WITH HTML CODE FROM SCREENSHOT TO CODE ``` " --------------------------------------------------------------------------------------------------

Michael Vandi

2023年12月6日

957

https://www.youtube.com/watch?v=nAoGNyYZ6CU

1:01:36

AI in Hollywood and Fine-Tuning Open-Source LLMs | Multimodal Weekly 29

In the 29th session of Multimodal Weekly, we featured two founders working with Generative AI and open-source AI. ✅ Russell Palmer, CEO & Co-Founder of CyberFilm AI, will give a presentation on how AI can benefit Hollywood and share his tool SAGA. - Connect with Russell here: https://www.linkedin.com/in/russellsapalmer/ - Check out CyberFilm AI: https://cyberfilm.ai/ - Try out SAGA: https://writeonsaga.com/ ✅ Michael Vandi, Founder and CEO of Addy AI, will give a presentation on fine-tuning open-source LLMs with LangDrive. - Connect with Michael: https://www.linkedin.com/in/michael-vandi/ - Check out Addy AI: https://addy-ai.com/ - Try out LangDrive: https://docs.langdrive.ai/ Timestamps: 02:40 Introduction 06:42 Russell starts 07:19 Act 1 12:20 Digital, 4k, VR, DVD, drones, green-screen... 14:40 Act 2 16:30 AI can be a force for good 18:10 Can AI make a good movie? 19:35 Why now? What's different? 23:55 Act 3 (GPT-4, DALL-E3, Midjourney, Pika, Runway) 29:54 AI + Human = Winner 32:25 A Live Demo of SAGA 38:30 Michael starts 40:45 Fine-tuning is hard 42:02 LangDrive makes fine-tuning easy 42:39 Let's dive into LangDrive docs 46:10 A Live Demo of LangDrive 48:45 Custom Data Loaders 49:35 Benefits of fine-tuning 51:25 Check out LangDrive on GitHub (https://github.com/addy-ai/langdrive) 52:25 Q&A for Michael Join the Multimodal Minds community to receive an invite for future webinars: https://discord.gg/Sh6BRfakJa

Twelve Labs

2024年1月25日

193

https://www.youtube.com/watch?v=B49WjGSnDcM