I built a website where you can easily fine-tune Llama 3.1 models

Easy Llama 3.1 Fine-Tuning

3.1-405B: an open source model to rival GPT-4o / Claude-3.5

I wrote this over the last couple of weeks using Rails 3, Twilio and a variety of Gems, putting in about 2 hours per day. I'd love any feedback from the HN community on any aspects of the site (design, API construction, documentation, marketing, pricing which is TBD). On a related note: I have to say that I'm really flying with Rails 3. Typically I roll with Django, but have been super impressed with the Rails 3. Feels really clean, and there are some Gems that make the more mundane parts of the webapp construction fun and agile: Haml/Sass, blueprint, Fancy-buttons, Devise. I'll probably…

Hey HN! I've been working on Muxamp for over a year now, and I decided that there's no point in keeping the source closed. (I never planned on making money from it.) I've learned a lot since I started, and I hope others can benefit from its release--and perhaps teach me a thing or two as well :)

Hey HN! After using a combination of Unsloth and Axolotl a lot, and finding it generally painful to figure out the right performance tuning for things like batch sizing and multi-GPU sharding, I wrote a small Python lib that sets up known-good LoRA training configurations for Llama 3.1 8B and 70B Instruct, and includes helpers for distilling from larger models or training on serverless finetuning platforms, and includes a walkthrough for distilling DeepSeek-R1 into a Llama 3.1 8B LoRA... But you can use it for pretty much any finetuning task, not just distilling large models!

Hi there, looking for feedback on my new project "Featherless.AI" The idea is to allow users to run all the models on hugging face instantly. Via the OpenAI API compatible endpoint. Why? Because its a real chore to download models and spin up GPUs, especially if you want to test multiple models. Not to mention GPUs cost multiple dollars an hour to rent. And if we want more people to use open source AI, we got to make it easier for them to try and play with all of them. So what if instead of spinning up dedicated GPUs per model (which is what every provider is doing) We can startup a LLM…