1-Click Deploy for Langchain LLM Agents from Google Colab to Web-App
We’ve just released Berri AI - a Python package https://github.com/ClerkieAI/berri_ai that makes it easy for developers to quickly deploy their LLM Agent from Google Colab to production (Web App and API Endpoint). Building LLM Apps can require working in online coding environments, like Colab, due to local environment limitations (e.g. running pytorch on older Macs). This can cause long dev cycles when deploying the app to production, as ported changes can only be tested once it is deployed after lengthy (>20min+) Docker builds. Berri lets you deploy directly from your…
In plain words
Berri AI is a Python package that enables developers to deploy Langchain LLM agents directly from Google Colab to production web applications and API endpoints. It streamlines the development process by eliminating lengthy Docker builds and the need to transfer code between environments. Developers install the package, import a function, and run deploy to share their LLM applications. The tool targets developers working in online coding environments who face constraints with local setups.
written from the facts on this page · September 2026
From the sources
In the maker’s words, at launch
We’ve just released Berri AI - a Python package https://github.com/ClerkieAI/berri_ai that makes it easy for developers to quickly deploy their LLM Agent from Google Colab to production (Web App and API Endpoint). Building LLM Apps can require working in online coding environments, like Colab, due to local environment limitations (e.g. running pytorch on older Macs). This can cause long dev cycles when deploying the app to production, as ported changes can only be tested once it is deployed after lengthy (>20min+) Docker builds. Berri lets you deploy directly from your coding environment, eliminating the need to download, transfer your code to a wrapper, deploy and debug. Taking an LLM app from your notebook and sharing it with people is made easy. How it works? Just install the package, import the function, and run deploy. At the end of the deploy (~10-15mins), you will get: A web app to interact with your agent https://agent-repo-35aa2cf3-a0a1-4cf8-834f-302e5b7fe07e-4524... An endpoint you can query https://agent-repo-35aa2cf3-a0a1-4cf8-834f-302e5b7fe07e-4524... is obama?" Want a more detailed walkthrough? Check out our loom - https://www.loom.com/share/fd4375b4a77f4ea7802369cb06a16d43 We’re still early so would love your feedback and opinions. Feel free to try us out for free – and if you need help building an agent / want a specific integration, just let us know!
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 17d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 27d ago · cactuscompute.com
