๐๐ข๐ On-Prem Agents with LangGraph Platform
Published about 1 month agoย โขย 2 min read
โ
โ
๐ Hey, AIM community!
โ
As we near the end of 2024, our team is looking back at all we've accomplished as a community this year. Thanks to all of you for learning ๐, building ๐, shipping ๐ข, and sharing ๐ with us at the open-source LLM Edge!
โ
We'll be rooting for you to take your AI career to the next level in 2025, and when you do, we hope you'll lean on us to amplify your story and showcase your best work.
โ
In this way, you'll help the AI Makerspace community achieve its mission of becoming the world's leading community for people who want to build, ship, and share production LLM applications.
โ
Here's to the bright future ahead for all of us! ๐ฅ
โ
As we settle in to take a short break between Christmas and the New Year, we'd like to wish all of you a very merry holiday season.
โ
See you in 2025 for more adventures from the LLM Edge!
โ
โ
๐ฎ A Few Predictions for 2025
Dr. Greg and The Wiz have spent the year learning, building, shipping, and sharing, and here's what they think you should look forward to in the new year!
Test-time Compute: We expect test-time compute to emerge as the new scaling axis for reasoning capabilities in LLMs. If this is a new term for you, investigate it now! [OpenAI Ref, Hugging Face Ref]
Synthetic Data Generation (SDG): Fully SDG models will outcompete non-SDG models on task-specific benchmarks. SDG will enhance the entire training stack, from pretraining to post-training. [Llama 3 models are already doing this Ref]
Multi-Agent Systems: Complex multi-agent systems will still not consistently generate business value in 2025. Single agents will be the sweet spot, and RAG will be table stakes.
Technical vs. Non-Technical: Prototypes will only get easier to build, even without formal coding experience. An app or website created by someone without any coding ability will go viral (e.g., App Store Top-100).
For our last event of the year, we explored on-prem with the LangGraph Platform! This was a newly released way to build agents within the LangChain ecosystem. The core components we need to understand within the LangChain ecosystem in 2025 are the LangGraph Server, LangGraph Studio, and the LangGraph CLI and Python SDK. We also discussed new releases, including LangGraph Command and LangGraph Interrupt! This was a great session that allowed us to wrap up the evolution of LangGraph this year!
The AI Makerspace team will take a break from YouTube Live events until the new year. You'll find us back in action starting Jan. 8, 2025!
๐ Around the Community!
๐ก Transformation Spotlight:Christos Magganas, a long-time AIM supporter and community member, shares his journey into Generative AI, including the apps he's built for companies and during hackathons. "For me, the Generative coding is the most exciting and valuable tool."
๐ฆพ LLM Engineering, Cohort 3 Graduates! Congratulations to #LLME3! We had a blast exploring LLM Engineering together, from the transformer to state-of-the-art Small Language Models (SLMs). We can't wait to see how these new AIM alumni build, ship, and share custom LLMs (and SLMs) in 2025!
Weekly Concepts, Code, and Community! Every Saturday you'll receive a detailed overview of one the latest tools or techniques from the open-source LLM edge.
Hey, AIM community! Join Dr. Greg and The Wiz as they cover Large Reasoning Models next Wednesday, Jan 18. With the release of o1, o3, and the new Gemini, everyone is talking about Chain-of-Thought Reasoning and Test-Time Compute. What are these things, anyway? And what are the implications for building production LLM applications with models like this in 2025 and beyond? Join the discussion live on Wed. at 10 AM PT! Last week, we dove into the latest RAG Evaluation metrics and RAGAS...
๐ Hey, AIM community! Dr. Greg and the Wiz will go on-prem with LangGraph next week! Join us for our last YouTube Live event before the New Year ๐! Last Wednesday, Dr. Greg and The Wiz guest spoke with Malikeh from Arcee on the SLM Show about the year in summary at the LLM Edge, and what to expect in 2025! We also explored vLLM! We learned that Virtual LLM helps us relieve memory bottlenecks when serving LLMs through PagedAttention, just like Virtual Memory relieves memory bottlenecks in...
๐ Hey, AIM community! Dr. Greg and the Wiz will unlock vLLM for you next week with a full breakdown of "Easy, fast, and cheap LLM serving for everyone." Last Wednesday, we explored AG2: AutoGen, Evolved with co-creator Qingyun Wu. The origin story was fascinating - from MathChat to going viral! AutoGen is all about conversations - which effectively constitute reasoning - by going full send on messages. ๐งฐ Resources ๐ง๐ซ Concepts: Slides ๐ง๐ป Code: CaptainAgent Notebook ๐ Paper: AutoGen The AutoGen...