zlacker

[parent] [thread] 6 comments
1. Camper+(OP)[view] [source] 2026-02-03 17:24:32
Please consider authoring a single, straightforward introductory-level page somewhere that explains what all the filename components mean, and who should use which variants.

The green/yellow/red indicators for different levels of hardware support are really helpful, but far from enough IMO.

replies(2): >>daniel+A4 >>segmon+R8
2. daniel+A4[view] [source] 2026-02-03 17:40:51
>>Camper+(OP)
Oh good idea! In general UD-Q4_K_XL (Unsloth Dynamic 4bits Extra Large) is what I generally recommend for most hardware - MXFP4_MOE is also ok
replies(1): >>Keats+Ju
3. segmon+R8[view] [source] 2026-02-03 17:57:00
>>Camper+(OP)
The green/yellow/red indicators are based on what you set for your hardware on huggingface.
◧◩
4. Keats+Ju[view] [source] [discussion] 2026-02-03 19:20:08
>>daniel+A4
Is there some indication on how the different bit quantization affect performance? IE I have a 5090 + 96GB so I want to get the best possible model but I don't care about getting 2% better perf if I only get 5 tok/s.
replies(1): >>mirekr+CJ
◧◩◪
5. mirekr+CJ[view] [source] [discussion] 2026-02-03 20:24:55
>>Keats+Ju
It takes download time + 1 minute to test speed yourself, you can try different quants, it's hard to write down a table because it depends on your system ie. ram clock etc. if you go out of gpu.

I guess it would make sense to have something like max context size/quants that fit fully on common configs with gpus, dual gpus, unified ram on mac etc.

replies(1): >>Keats+1S
◧◩◪◨
6. Keats+1S[view] [source] [discussion] 2026-02-03 21:06:00
>>mirekr+CJ
Testing speed is easy yes, I'm mostly wondering about the quality difference between Q6 vs Q8_K_XL for example.
replies(1): >>daniel+Cx1
◧◩◪◨⬒
7. daniel+Cx1[view] [source] [discussion] 2026-02-04 00:57:00
>>Keats+1S
I haven't done benchmarking yet (plan to do them), but it should be similar to our post on DeepSeek-V3.1 Dynamic GGUFs: https://unsloth.ai/docs/basics/unsloth-dynamic-2.0-ggufs
[go to top]