danielhanchen

@[email protected]
0 Followers
0 Following
5 Posts
Unsloth github.com/unslothai/unsloth - finetune Llama 2x faster + use 70% less VRAM

1. Used to work at NVIDIA RAPIDS cuML

2. Discord: https://discord.gg/unsloth

3. Github: https://github.com/danielhanchen

4. Twitter / X: x.com/danielhanchen

5. Email: my handle @ gmail.com

6. Bug fixes for Gemma: https://news.ycombinator.com/item?id=39671146

7. Bug fixes for Gradient Accumulation: https://x.com/danielhanchen/status/1846235913443262891?lang=en
This account is a replica from Hacker News. Its author can't see your replies. If you find this service useful, please consider supporting us via our Patreon.

Officialhttps://
Support this servicehttps://www.patreon.com/birddotmakeup
Show threaddanielhanchenMar 18

If you can try again we just updated the process sorry! We did a new pypi release:

curl -LsSf https://astral.sh/uv/install.sh | sh

uv venv unsloth_studio --python 3.13

source unsloth_studio/bin/activate

uv pip install unsloth==2026.3.7 --torch-backend=auto

unsloth studio setup

unsloth studio -H 0.0.0.0 -p 8888

Show threaddanielhanchenMar 18

Hey! Our primary objective for now is to provide the open source community with cool and useful tooling - we found closed source to be much more popular because of better tooling!

We have much much in the pipeline!!

Show threaddanielhanchenMar 17
Yep - uv is a better fit - and you get parallel downloads as well
Show threaddanielhanchenMar 17
Yep uv pip install unsloth works as well - we probably should have just made that the default - in fact Unsloth makes its own venv using UV if you have it dynamically
Show threaddanielhanchenMar 17

Hey we're still working on making installation much better - appreciate the feedback!

We come from Python land mainly so packaging and distribution is all very new to us - homebrew will definitely be next!