正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

I just created my own OCR app using Llama 3.2 vision! Upload an image, and it converts it into structured markdown using Llama 3.2 multimodal! Here's what I used: - Ollama for serving Llama 3.2 vision locally. - Streamlit for the UI. Everything in just 50 lines of code!... show more

Avi Chawla

67,098 subscribers

131,139 次观看 • 1 年前 •via X (Twitter)

教育科学技术

Anya Rossi• Live Now

Private livecam show

11 条评论

Avi Chawla 的头像

Avi Chawla1 年前

Code:

Matias Perelli 的头像

Matias Perelli1 年前

I recorded a 31-minute video on how 7-8 figure eCom brands can add $143k using 3 Klaviyo "tweaks". Watch the exact method we used recently with a client. Click below.

Ivan Fioravanti ᯅ 的头像

Ivan Fioravanti ᯅ1 年前

Show me the code 🤣

Avi Chawla 的头像

Avi Chawla1 年前

Here you go:

hoka 的头像

hoka1 年前

Bro why

Avi Chawla 的头像

Avi Chawla1 年前

Akshay and I are co-founders of the same newsletter. So we work together.

remy 的头像

remy1 年前

Not as pretty but I actually did one last night to help someone. One shot ~ich with Cursor. :-)

Paul Lemaistre 的头像

Paul Lemaistre1 年前

Why not use stepfun-ai/GOT-OCR2_0? As far as I know it’s SOTA

Jayson Gent 的头像

Jayson Gent1 年前

@akshay_pachaar Awesome! Didn't even know a vision model was available. I need more vram..

Hemant C. Sharma 的头像

Hemant C. Sharma1 年前

Can anyone give me a solution that will run locally on a mobile app, without needing to connect. Thank you in advance

jatin 的头像

jatin1 年前

Product idea: Integrate with Concur and other expense filing platforms. Extract this data as a json and pass it on to their API to automatically file expenses.

相关视频

I just built my own OCR app powered by Google's Gemma 3. It runs 100% locally on my computer using Ollama and extracts text from images as structured markdown. The code is open-source, and you can find it in the next tweet.

I just built my own OCR app powered by Google's Gemma 3. It runs 100% locally on my computer using Ollama and extracts text from images as structured markdown. The code is open-source, and you can find it in the next tweet.

Akshay 🚀

478,020 次观看 • 1 年前

$I just added the new Llama 3.2 1B and 3B models to LitGPT, the open-source LLM library I help develop (focused on efficiency and code readability). LitGPT allows you to fine-tune and use these models on the cloud or a laptop. So, if you are looking for something to play with this weekend: # 1) Finetune the model litgpt finetune_lora meta-llama/Llama-3.2-1B \ --data JSON \ --data.json_path my_custom_dataset.json \ --train.epochs 1 \ --out_dir out/llama-3.2-finetuned \ --precision bf16-true # 2) Chat with the model litgpt chat out/llama-3.2-finetuned/final # 3) Serve the model via an API endpoint litgpt serve out/llama-3.2-finetuned/final$

I just added the new Llama 3.2 1B and 3B models to LitGPT, the open-source LLM library I help develop (focused on efficiency and code readability). LitGPT allows you to fine-tune and use these models on the cloud or a laptop. So, if you are looking for something to play with this weekend: # 1) Finetune the model litgpt finetune_lora meta-llama/Llama-3.2-1B \ --data JSON \ --data.json_path my_custom_dataset.json \ --train.epochs 1 \ --out_dir out/llama-3.2-finetuned \ --precision bf16-true # 2) Chat with the model litgpt chat out/llama-3.2-finetuned/final # 3) Serve the model via an API endpoint litgpt serve out/llama-3.2-finetuned/final

Sebastian Raschka

65,529 次观看 • 2 年前

I just created an agentic-workflow to automatically write and publish content for me! It's powered by CrewAI Flows and Llama 3.2, running 100% locally. Tech stack: - CrewAI to build an agentic workflow - FireCrawl for web scraping - Typefully for scheduling Here's how it works: - You provide a link to a website. - It scrapes and saves the data as markdown. - A router triggers the desired Crew of agents. - The Crew prepares a ready-to-publish draft. - Finally, use Typefully to post it to your socials. Totally hands-off and 100% automated! In this video, I provide a deep dive into how it actually works! Find the link to all the code in the next tweet! Enjoy the video! 🥂

I just created an agentic-workflow to automatically write and publish content for me! It's powered by CrewAI Flows and Llama 3.2, running 100% locally. Tech stack: - CrewAI to build an agentic workflow - FireCrawl for web scraping - Typefully for scheduling Here's how it works: - You provide a link to a website. - It scrapes and saves the data as markdown. - A router triggers the desired Crew of agents. - The Crew prepares a ready-to-publish draft. - Finally, use Typefully to post it to your socials. Totally hands-off and 100% automated! In this video, I provide a deep dive into how it actually works! Find the link to all the code in the next tweet! Enjoy the video! 🥂

Akshay 🚀

98,456 次观看 • 1 年前

"Introducing Multimodal Llama 3.2": As promised two weeks ago, here's the short course on Meta's latest open model! This short course is created with Meta and taught by Amit Sangani, Director of AI Partner Engineering at Meta. Meta’s Llama family of models is leading the way in open models, allowing anyone to download, customize, fine-tune, or build new applications on top of them. Learn about the vision capabilities of the Llama 3.2, and use it for image classification, prompting, tokenization, tool-calling. You'll also learn about the open-source Llama stack, which gives building blocks for many different stages of the LLM application life cycle. In detail, you’ll: - Learn what are the features of Meta's four newest models, and when to use which Llama model. - Learn best practices for multimodal prompting, with applications to advanced image reasoning, illustrated by many examples: Understanding errors on a car dashboard, adding up the total of photographed restaurant receipts, grading written math homework. - Use different roles—system, user, assistant, ipython—in the Llama 3.1 and 3.2 models and the prompt format that identifies those roles. - Understand how Llama uses the tiktoken tokenizer, and how it has expanded to a 128k vocabulary size that improves encoding efficiency and multilingual support. - Learn how to prompt Llama to call built-in and custom tools (functions) with examples for web search and solving math equations. - Learn about Llama Stack, a standardized interface for common toolchain components like fine-tuning or synthetic data generation, useful for building agentic applications. By the end of this course, you’ll be equipped to build out new applications with the new Llama 3.2. Thank you to Ahmad Al-Dahle, Amit Sangani, and the whole AI at Meta team AI at Meta for all the hard work on Llama 3.2 — we’re excited to make these open models even more accessible to more developers with this new course! Please sign up here!

"Introducing Multimodal Llama 3.2": As promised two weeks ago, here's the short course on Meta's latest open model! This short course is created with Meta and taught by Amit Sangani, Director of AI Partner Engineering at Meta. Meta’s Llama family of models is leading the way in open models, allowing anyone to download, customize, fine-tune, or build new applications on top of them. Learn about the vision capabilities of the Llama 3.2, and use it for image classification, prompting, tokenization, tool-calling. You'll also learn about the open-source Llama stack, which gives building blocks for many different stages of the LLM application life cycle. In detail, you’ll: - Learn what are the features of Meta's four newest models, and when to use which Llama model. - Learn best practices for multimodal prompting, with applications to advanced image reasoning, illustrated by many examples: Understanding errors on a car dashboard, adding up the total of photographed restaurant receipts, grading written math homework. - Use different roles—system, user, assistant, ipython—in the Llama 3.1 and 3.2 models and the prompt format that identifies those roles. - Understand how Llama uses the tiktoken tokenizer, and how it has expanded to a 128k vocabulary size that improves encoding efficiency and multilingual support. - Learn how to prompt Llama to call built-in and custom tools (functions) with examples for web search and solving math equations. - Learn about Llama Stack, a standardized interface for common toolchain components like fine-tuning or synthetic data generation, useful for building agentic applications. By the end of this course, you’ll be equipped to build out new applications with the new Llama 3.2. Thank you to Ahmad Al-Dahle, Amit Sangani, and the whole AI at Meta team AI at Meta for all the hard work on Llama 3.2 — we’re excited to make these open models even more accessible to more developers with this new course! Please sign up here!

Andrew Ng

131,846 次观看 • 2 年前

i have built the most innovative AI community in the world... here's how i used AI agents to build it for me: - the community lives on Whop, and i control it entirely from their CLI - every section is its own app, so i push updates to the lessons and ship REAL software inside the community - my agent has full vision on who's using what and where engagement peaks, so the upsell sits exactly where it should i firmly believe this is the future of education

i have built the most innovative AI community in the world... here's how i used AI agents to build it for me: - the community lives on Whop, and i control it entirely from their CLI - every section is its own app, so i push updates to the lessons and ship REAL software inside the community - my agent has full vision on who's using what and where engagement peaks, so the upsell sits exactly where it should i firmly believe this is the future of education

Machina

14,766 次观看 • 1 个月前

I was able to create a prompt that helps me generate bulks of images using flow agent. this is a 61 image I generated in one go just by using one character image and it used the style across every 61 prompt And everything was done on mobile.. Yt automators this is for you😎

I was able to create a prompt that helps me generate bulks of images using flow agent. this is a 61 image I generated in one go just by using one character image and it used the style across every 61 prompt And everything was done on mobile.. Yt automators this is for you😎

Peter Buildweb | Al

21,243 次观看 • 3 个月前

New short course: Prompt Engineering with Llama 2, built in collaboration with Meta AI at Meta, and taught by Amit Sangani! Meta's Llama 2 has been game-changing for AI. Building with open source lets you control your own data, scrutinize errors, update (or not) the models as you please, and work alongside the global community advancing open models. Llama isn't a single model, it's a collection of models. In this course, you'll: - Learn the differences between different Llama 2 flavors, and when to use each. - Prompt the Llama chat models -- you'll also see how Llama's instruction tags work -- so they can help you with day-to-day tasks, like writing or summarization. - Use advanced prompting, like few-shot prompting for classification, and chain-of-thought prompting for solving logic problems. - Use specialized models in the Llama collection for specific tasks, like Code Llama to help you write, analyze, and improve code, and Llama Guard, which checks prompts and model responses for harmful content. The course also touches on how to run Llama 2 locally on your own computer. I hope you’ll take this course and try out these powerful, open models!

New short course: Prompt Engineering with Llama 2, built in collaboration with Meta AI at Meta, and taught by Amit Sangani! Meta's Llama 2 has been game-changing for AI. Building with open source lets you control your own data, scrutinize errors, update (or not) the models as you please, and work alongside the global community advancing open models. Llama isn't a single model, it's a collection of models. In this course, you'll: - Learn the differences between different Llama 2 flavors, and when to use each. - Prompt the Llama chat models -- you'll also see how Llama's instruction tags work -- so they can help you with day-to-day tasks, like writing or summarization. - Use advanced prompting, like few-shot prompting for classification, and chain-of-thought prompting for solving logic problems. - Use specialized models in the Llama collection for specific tasks, like Code Llama to help you write, analyze, and improve code, and Llama Guard, which checks prompts and model responses for harmful content. The course also touches on how to run Llama 2 locally on your own computer. I hope you’ll take this course and try out these powerful, open models!

Andrew Ng

162,875 次观看 • 2 年前

I just vibe coded an app powered by an n8n agent, using the . The app will help me do research and prep for podcasts that I produce. Here's how the app works: - I type in the episode topic or topics - The App connects with my n8n agent that researches the topic, relevant news, and what people are saying about it - It also finds relevant links, and tweets - It sends the script / outline email to the team - And then it shows me what it did on the app

I just vibe coded an app powered by an n8n agent, using the . The app will help me do research and prep for podcasts that I produce. Here's how the app works: - I type in the episode topic or topics - The App connects with my n8n agent that researches the topic, relevant news, and what people are saying about it - It also finds relevant links, and tweets - It sends the script / outline email to the team - And then it shows me what it did on the app

Emily Lambert

57,603 次观看 • 1 年前

Turn any GitHub repository into LLM-ready text! Simply replace "hub" with "ingest" in a GitHub URL and receive a prompt-friendly text ingest for LLMs. Gitingest is 100% open-source and provides: - Directory structure - A brief summary of the project - The entire content as LLM-ready text Plus, it comes with a nice python package and you can run the UI locally! Stay tuned, I'm working on something really cool with this!✨ Link to the GitHub repo in the next tweet! _____ Find me → Akshay 🚀 ✔️ For more insights and tutorials on ML and AI Engineering!

Turn any GitHub repository into LLM-ready text! Simply replace "hub" with "ingest" in a GitHub URL and receive a prompt-friendly text ingest for LLMs. Gitingest is 100% open-source and provides: - Directory structure - A brief summary of the project - The entire content as LLM-ready text Plus, it comes with a nice python package and you can run the UI locally! Stay tuned, I'm working on something really cool with this!✨ Link to the GitHub repo in the next tweet! _____ Find me → Akshay 🚀 ✔️ For more insights and tutorials on ML and AI Engineering!

Akshay 🚀

191,577 次观看 • 1 年前

The new Screenshot to Code from Vercel v0 is insane. I just took a screenshot of a pricing table and it was converted into code using AI in seconds! It even outputs the code and allows you to modify any specific parts.

The new Screenshot to Code from Vercel v0 is insane. I just took a screenshot of a pricing table and it was converted into code using AI in seconds! It even outputs the code and allows you to modify any specific parts.

Alvaro Cintas

378,069 次观看 • 2 年前

Markdown is the new programming language. But writing markdown sucks. So I created a beautiful app to write and edit markdown files with agents, locally. It's called Skribe, an open-source, beautiful markdown editor and you can try it for free today. Just: - Open your codebase - Only markdown files are shown - Edit and write markdown alongside Claude Code or Codex within a beautiful editor

Markdown is the new programming language. But writing markdown sucks. So I created a beautiful app to write and edit markdown files with agents, locally. It's called Skribe, an open-source, beautiful markdown editor and you can try it for free today. Just: - Open your codebase - Only markdown files are shown - Edit and write markdown alongside Claude Code or Codex within a beautiful editor

Chris Ashby

11,258 次观看 • 3 个月前

I just used Luma's text-to-3D to generate a DeLorean from Back to the Future and placed it in my living room using the Vision Pro! The future is wild. 🤩

I just used Luma's text-to-3D to generate a DeLorean from Back to the Future and placed it in my living room using the Vision Pro! The future is wild. 🤩

Ammaar Reshi

265,563 次观看 • 2 年前

Introducing DeepSeek R1:1.5b Running Locally in Cursor! In less than 4 minutes, I set up the DeepSeek R1:1.5b model, download it, and run it locally to seamlessly work with code in Cursor The video is shown in real-time, with the model impressively running on an Intel Core i5 😱 using Ollama and the CodeGPT extension 👇

Introducing DeepSeek R1:1.5b Running Locally in Cursor! In less than 4 minutes, I set up the DeepSeek R1:1.5b model, download it, and run it locally to seamlessly work with code in Cursor The video is shown in real-time, with the model impressively running on an Intel Core i5 😱 using Ollama and the CodeGPT extension 👇

Daniel San

584,100 次观看 • 1 年前

I built an app with Claude Opus 4.5, then animated it with this new tool I found... This new iOS design workflow let me animate a character for my onboarding. Here it is: 1. build the app using Claude Code within Vibecode 2. entered image of my mascot into 3. copied animation code from there back into vibecode

I built an app with Claude Opus 4.5, then animated it with this new tool I found... This new iOS design workflow let me animate a character for my onboarding. Here it is: 1. build the app using Claude Code within Vibecode 2. entered image of my mascot into 3. copied animation code from there back into vibecode

Emily Lambert

292,874 次观看 • 9 个月前

I took Clicky and I made it 5x Faster So I saw that Farza 🇵🇰🇺🇸 was using uses Claude's vision to find UI elements on screen, send a screenshot, wait for coordinates back. It works, but it's slow. I replaced that with OmniParser V2 by Microsoft which is a local Object detection model trained specifically on UI elements. It runs on-device, detects every button, menu, and icon in 400ms, and gives me pixel-precise coordinates. No API call, no latency, no cost. The green highlights you see around the UI elements is the detection overlay and you can see as I am switching the screen it takes no time to detect and highlight which is pretty neat! With Local models improving by the day, Its the right direction for applications like these. Next up: video-synced tutorials where a YouTube tutorial pauses and waits for you to perform each action in the real app. I am not stopping!

I took Clicky and I made it 5x Faster So I saw that Farza 🇵🇰🇺🇸 was using uses Claude's vision to find UI elements on screen, send a screenshot, wait for coordinates back. It works, but it's slow. I replaced that with OmniParser V2 by Microsoft which is a local Object detection model trained specifically on UI elements. It runs on-device, detects every button, menu, and icon in 400ms, and gives me pixel-precise coordinates. No API call, no latency, no cost. The green highlights you see around the UI elements is the detection overlay and you can see as I am switching the screen it takes no time to detect and highlight which is pretty neat! With Local models improving by the day, Its the right direction for applications like these. Next up: video-synced tutorials where a YouTube tutorial pauses and waits for you to perform each action in the real app. I am not stopping!

Milind S

73,599 次观看 • 5 个月前

Vibe code powerful photo editing apps using the new Gemini 2.5 Flash Image! 🚀 I just made an app powered by nano banana 🍌 that takes your photo and turns it into polaroids from iconic decades. Export albums, share with friends, shake to regenerate! Demo + code below!

Vibe code powerful photo editing apps using the new Gemini 2.5 Flash Image! 🚀 I just made an app powered by nano banana 🍌 that takes your photo and turns it into polaroids from iconic decades. Export albums, share with friends, shake to regenerate! Demo + code below!

Ammaar Reshi

83,287 次观看 • 1 年前

Grok Code generates the most aesthetic UI/UX Design Built a stunning carousel with mesmerizing hover effects & glows - just 2 prompts in Kilo Code, and it nailed every detail Grok Code created Frosty esthetics - snowflakes, glassy vibes, everything I asked for, for a live feel Animated and interactive elements, clickable 𝕏 profile and links everything perfectly connected, super smooth, and insanely responsive Grok Code also created a stunning side-show, smooth transitions, glowing borders with Grok Imagine images It is straight-up a powerhouse for visual code and pure UI magic

X Freeze

1,104,236 次观看 • 1 年前

Is Windsurf better than Cursor? impossible to say, but... it does one thing better: soooo... I spent an hour testing windsurf... and this is what i created... and app called "boop-boop" lol... one thing windsurf does is it searches the internet. If you've been watching my videos you know that i go to perplexity all the time to get code examples and docs. but with windsurf you can just say, "search the replicate top image gen models and add them to the chat we just created" It will search the internet, find the right model then write the code. This is great for replicate and openrouter. You can have one api key in your .env file then just tell windsurf to search for docs for [insert ai model] and it will just add it to your app. Impressed tbh.

Is Windsurf better than Cursor? impossible to say, but... it does one thing better: soooo... I spent an hour testing windsurf... and this is what i created... and app called "boop-boop" lol... one thing windsurf does is it searches the internet. If you've been watching my videos you know that i go to perplexity all the time to get code examples and docs. but with windsurf you can just say, "search the replicate top image gen models and add them to the chat we just created" It will search the internet, find the right model then write the code. This is great for replicate and openrouter. You can have one api key in your .env file then just tell windsurf to search for docs for [insert ai model] and it will just add it to your app. Impressed tbh.

Riley Brown

31,043 次观看 • 1 年前