This is really impressive. Can you say a bit about the underlying process? I'm guessing this is post-training qantization? Isn't PTQ also resource-intensive? (Ie might not work on any machine)
does this work similar to airllm? i am wondering how it would handle something like quantizing kimi k3 on a budget of 8 gbs, or is that something you are not attempting to solve yet?
This is interesting. I wonder how it could work with something like https://github.com/JustVugg/colibri.
I gotta laugh at some of the models it suggests, for example:
> AnkitAI/Parable-Qwen3-4B-Claude-Fable-5-GGUF
you’re telling me you managed to fit Fable 5 into just 4B?
Fyi that model name to me reads
Qwen3 4b params distilled/trained with fable 5
Reminds me of https://github.com/AlexsJones/llmfit
LLMFit tells you what can run on something. I built something quite similar to their search into Shoehorn now.
This is really impressive. Can you say a bit about the underlying process? I'm guessing this is post-training qantization? Isn't PTQ also resource-intensive? (Ie might not work on any machine)
The project name is perfect!
does this work similar to airllm? i am wondering how it would handle something like quantizing kimi k3 on a budget of 8 gbs, or is that something you are not attempting to solve yet?
Yes that is exactly what this does.
Could you explain what happens when you try to shoehorn a 2.4T parameter model into a 24gb m4 mac?
Wondering the same thing but for 48gb M5 Max.
extreme divergence would be my guess
tried it out but based on the model sizing result i got i got an insufficient memory error when the server started running
If you could post an issue if you still have the error around that would be awesome.