Fine-tune generative models in 1 line of code
Hey HN, excited to release Blueprint into public beta. Blueprint is a general fine-tuning and model serving API built for developers. Fine-tuning models is an extremely powerful way to improve performance on a specific task without needing to collect prohibitively large amounts of data. With Blueprint you can kick off fine-tuning jobs for various open source models like Stable Diffusion and soon Flan-T5 using a Python SDK. We'll also deploy your fine-tuned models onto serverless GPUs so that you aren't paying for idle GPU time. We scale the models up when you need to serve requests and have…
In plain words
Blueprint is an API that lets developers fine-tune open source generative models like Stable Diffusion and Flan-T5 with a single line of Python code. It handles model training and deploys fine-tuned models on serverless GPUs, automatically scaling up or down based on traffic so users only pay for active requests. The service is designed for developers who want to customize models for specific tasks without collecting large datasets.
written from the facts on this page · September 2026
From the sources
In the maker’s words, at launch
Hey HN, excited to release Blueprint into public beta. Blueprint is a general fine-tuning and model serving API built for developers. Fine-tuning models is an extremely powerful way to improve performance on a specific task without needing to collect prohibitively large amounts of data. With Blueprint you can kick off fine-tuning jobs for various open source models like Stable Diffusion and soon Flan-T5 using a Python SDK. We'll also deploy your fine-tuned models onto serverless GPUs so that you aren't paying for idle GPU time. We scale the models up when you need to serve requests and have put a ton of engineering work into faster cold starts. We'll also autoscale replicas of your model if your model is receiving a lot of traffic. Give it a shot — every new account gets a few hours of GPU credits. For support and feedback, join our Discord here: https://discord.gg/9pcXqWgB3g
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
Launched alongside, March 2023
the whole month →
- BI
I'm a big fan of the BBC podcast In Our Time -- and (like most people) I've been playing with the OpenAI APIs. In Our Time has almost 1,000 episodes on everything from Cleopatra to the evolution of teeth to plasma physics, all still available, so it's my starting point to learn about most topics. But it's not well organised. So here are the episodes sorted by library code. It's fun to explore. Web scraping is usually pretty tedious, but I found that I could send the minimised HTML to GPT-3 and get (almost) perfect JSON back: the prompt includes the Typescript definition. At the same time I…
AI · 2023 · genmon.github.io