Yea, I actually tried it out last time we had one of these threads. It's undeniably easy to use, but it is also very opinionated about things like the directory locations/layouts for various assets. I don't think I managed to get it to work with a simple flat directory full of pre-downloaded models on an NFS mount to my NAS. It also insists on re-downloading a 3GB model every time it is launches, even after I delete the model file. I probably have to just sit down and do some Googleing/searching in order to rein the software in and get it to work the way I want it to on my system.
Oh my apologies I didn't respond - if only HN had a notifier haha
Oh yes we added a custom folder button which can pull .gguf files for now from any folder - it supports LM Studio and Ollama ones - but afreed it's still a mess.
One of the goals is to somehow quick search for .gguf folders, and add recommended folders - we currently have folders for Ollama and LM Studio for eg
Sadly doesn't support fine tuning on AMD yet which gave me a sad since I wanted to cut one of these down to be specific domain experts. Also running the studio is a bit of a nightmare when it calls diskpart during its install (why?)
Apologies as well didn't reply sooner - Studio supports AMD out of the box now! We worked with AMD to make it work! One thing that is still missing is pre-compiled AMD ROCM binaries, which we're trying to see if we can integrate that.
Interesting on diskpart - let me check and get back to you [EDIT] - visual studio build tools, python 3.13, git, cmake, node.js are all msi-based installers - so these are likely the culprits on using diskpart - essentially MSI installers check if there's enough disk space before installing items
Thanks for that. Did you notice that the unsloth/unsloth docker image is 12GB? Does it embed CUDA libraries or some default models that justifies the heavy footprint?
Hey so sorry didn't reply sooner - yes the docker used to be I think 4-8GB ish since CUDA sadly itself is 4GB I think, and PyTorch takes the rest. So unfortunately the Unsloth Docker image has ballooned due to this. We tried reducing it as much as possible, but it's hard :( https://hub.docker.com/r/vllm/vllm-openai/tags for eg is around 11GB ish, ad we're 13.6GB ish.
We'll try our best to compress it more, but it's tough
I applaud that you recently started providing the KL divergence plots that really help understand how different quantizations compare. But how well does this correlate with closed loop performance? How difficult/expensive would it be to run the quantizations on e.g. some agentic coding benchmarks?
Hey! Sorry for not replying sooner - yes we'll keep publishing more KLD - sadly some are saying we are "optimizing" for KLD now since we posted so many haha - but the whole purpose of quantization is to match the BF16 logits as much as possible whilst reducing disk space (ie reduce KLD).
In general so this is funny and a quirk of quantization - sometimes 8bit, 4bit models do BETTER on downstream benchmarks (SWE Bench for eg), since sometimes rounding can actually somehow act as a "regularization" method (this is just my hunch).
So KLD isn't that expensive, since we leverage the trick of causal attention - since causal attention is lower triangular, we can do 1 forward pass on the enter text (say 2048 tokens), and you attain logits for the prediction for every token's position - so this is O(N^2).
However coding benchmarking require actual inference, and cannot use the causal attention trick, and it's best to run them 10 times since temperature = 1.0 is not deterministic - and take an average. We plan to maybe do something like https://marginlab.ai/trackers/claude-code/, which takes a random sample and does it over time.
Hey sorry on the delay - we just added API support, so you can access a remote server - it includes optional python, tool call, bash and web search support if you enable them.
For SSH - we haven't yet done that - for now we have a SHA256 encryption approach, but it's not SSH yet. HTTPS will also sadly have to be the end user's setup process as well - we plan to make it better soon!
Appreciate what y'all do! We were slacking about how many HGX-B300 it would take to run Kimi and it looks like we could actually fit 2-3 Kimis on a single HGX.
Sorry on the delay - so it installs https://github.com/Blaizzy/mlx-vlm and other components and sets up the commands - you don't need to use it but we thought it might be easier for folks
Oh hey - we're actually the 4th largest distributor of OSS AI models in GB downloads - see https://huggingface.co/unsloth
https://unsloth.ai/docs/basics/unsloth-dynamic-2.0-ggufs is what might be helpful. You might have heard 1bit dynamic DeepSeek quants (we did that) - not all layers can be 1bit - important ones are in 8bit or 16bit, and we show it still works well.
Yes this is fair - we try our best to communicate issues - I think we're mostly the only ones doing the communication that model A or B has been fixed etc.
We try our best as model distributors to fix them on day 0 or 1, but 95% of issues aren't our issues - as you mentioned it's the chat template or runtime etc
I have to ask - what do you run locally on your laptop (model, backend, and agentic cli)?
Feature request:
A leader board with filtering so you can enter your machine specs and it will sort all models along with all the various quantisation and then rank them all - because so far model ranking site either don’t include all available quants, don’t compare apples to apples (ie was one model tested with Claude code while another benchmark done with opencode) etc
Oh - and as bonus, scoring also ranked by which agentic CLI :)
1. Auto best official parameters set for all models
2. Auto determines the largest quant that can fit on your PC / Mac etc
3. Auto determines max context length
4. Auto heals tool calls, provides python & bash + web search :)
reply