LlamaIndex
153,231 followers
- Report this post
Big news! The open source Nomic embed can now be run fully locally!Not only can you specify totally local embeddings, but with their new dynamic inference mode, you can optimize for embedding latency and get the best of both local and remote embeddings -- the library will automatically select local embeddings when they are faster, and remote embeddings when they're not!Their blog post: https://lnkd.in/g6qVX9ipA demo notebook: https://lnkd.in/gMcN7Fmq
69
3 Comments
Olivier Mills
Baobab Tech | AI solutions for social impact | Founder of WASH AI
7h
- Report this comment
Who is serving it as low-latency serverless inference .. or via vector-database edge functions Supabase? Fireworks AI Anyscale Vercel ?
1Reaction
Jin W.
Generative AI Prompt Curator | Pharmacist | Connector of people across technology, healthcare and finance
8h
- Report this comment
Very cool
1Reaction 2Reactions
Boy_I - B.
@CHANELDB.IO - DATA on peer-to-peer secure channels
4h
- Report this comment
1Reaction
To view or add a comment, sign in
More Relevant Posts
-
LlamaIndex
153,231 followers
- Report this post
Automate real-world tasks with LlamaIndex and MultiOn!MultiOn is an AI agents platform that works with the web to get real things done by connecting to the Internet through your Chrome web browser and acting on your behalf.Check out our demo where it searches through your inbox, reads an email, and responds to it, all through code: https://lnkd.in/gwG3Fdtz
43
2 Comments
Like CommentTo view or add a comment, sign in
-
LlamaIndex
153,231 followers
- Report this post
Build a Full-Stack Job Search Assistant with Koyeb, MongoDB, and LlamaIndex 🧑💼🔎This is a comprehensive end to end tutorial by Rishi Raj Jain on building a RAG-powered assistant that streams its response in real-time but can also continuously update its internal collection of jobs. At the end of the tutorial, deploy to a web service! 🌐1. Setup MongoDB Atlas, seed it with data2. Create a nextjs application3. Create the shadcn/UI/markdown components4. Define the application routes, including job indexing endpoint and chat endpoint5. Deploy to KoyebA must-have resource if you’re an application developer looking for an e2e resource on building a web app around RAG. Check out the full blog here: https://lnkd.in/gGC6AhVU
104
1 Comment
Like CommentTo view or add a comment, sign in
-
LlamaIndex
153,231 followers
- Report this post
Use batch inference to pre-process your data for GenAI applications!Not every LLM query needs to be real-time; sometimes you have a large set of data that can benefit from being pre-processed by an LLM, making possible new kinds of analysis and querying.Our integration with MyMagic AI makes it easy to batch-process data, and this quick tutorial shows you how to then use the processed data to improve inference later.Check out their guest post on our blog: https://lnkd.in/gk9YtUJA
151
4 Comments
Like CommentTo view or add a comment, sign in
-
LlamaIndex
153,231 followers
- Report this post
We’re excited to release our webinar featuring OpenDevin - an open-source platform for building autonomous AI engineers. Robert Brennan, one of the lead maintainers, gives a fantastic walkthrough of the platform, and we discuss some high-level insights for building agents:- Have specialized “micro”-agents that can perform different tasks with reasoning loops- Have a higher-level delegator agent that can pick the right micro-agents, and have its own state machine to write/validate code.The result is a powerful platform that crushes SWE-Bench Lite benchmarks, and gives the Devin-like experience where the autonomous worker can write code with only sporadic human intervention. Video: https://lnkd.in/gT__uwCgOpenDevin repo: https://lnkd.in/dDRky3AcOpenDevin: https://lnkd.in/gUFMWFsx
205
3 Comments
Like CommentTo view or add a comment, sign in
-
LlamaIndex
153,231 followers
- Report this post
✨ Launching today! ✨Securely run LLM-generated code in a sandbox using Azure Container Apps dynamic sessions!When building an agent, it can be very helpful to provide it a tool that uses code to perform calculations or other tasks that LLMs are not natively good at.LLMs are increasingly capable at generating code to solve complex problems, but that creates risks -- what if the code is dangerous or insecure? Now the same technology ChatGPT uses to execute generated code securely is now available to everyone, and integrated directly into LlamaIndex.To use it, spin up a session pool on Azure, then provide our new AzureCodeInterpreterTool to your agent, and that's it!https://lnkd.in/gSEFajAn
121
4 Comments
Like CommentTo view or add a comment, sign in
-
LlamaIndex
153,231 followers
- Report this post
Complex Document RAG with GPT-4o 📑GPT-4o’s multimodal capabilities means that it’s great at parsing complex PDFs/slide decks with background images, irregular layouts, charts into semi-structured markdown with text and tables.We’ve natively integrated GPT-4o with LlamaParse letting you use it as a key part in the RAG ETL step. Our cookbook below shows you how to parse the 2019 Tesla impact report deck into text representations, and ask questions over chart data in the original deck 🔥Bonus 💫: GPT-4o on LlamaParse is now $0.03 per page! (20x reduction from before). Notebook: https://lnkd.in/grwUVr-GSignup for LlamaParse: https://lnkd.in/gi8dxGntLlamaParse repo: https://lnkd.in/g3UmUkcD
584
25 Comments
Like CommentTo view or add a comment, sign in
-
LlamaIndex
153,231 followers
- Report this post
Our TypeScript docs just got an upgrade! Head over to ts.llamaindex.ai to see our four new starter tutorials as well as our extensive new step-by-step guide to building agents in TypeScript: https://lnkd.in/gC7h5EzX
42
Like CommentTo view or add a comment, sign in
-
LlamaIndex
153,231 followers
- Report this post
Reminder: next Tuesday we're having our first ever in-person meetup at our San Francisco HQ! Come meet the team and hear great talks from Jerry Liu and our partners Tryolabs and Activeloop!RSVP here: https://lnkd.in/d3HXb3wF
39
2 Comments
Like CommentTo view or add a comment, sign in
-
LlamaIndex
153,231 followers
- Report this post
Paris residents! Catch our own Pierre-Loic Doulcet at the (amazingly-named) PizzerIA on Thursday giving a talk on advanced retrieval-augmented generation techniques!https://lnkd.in/g3SJgWAf
20
Like CommentTo view or add a comment, sign in
153,231 followers
View Profile
Follow