π Model Description
base_model: cerebras/GLM-4.6-REAP-218B-A32B language:
- en
- glm
- MOE
- pruning
- compression
About
weighted/imatrix quants of https://huggingface.co/cerebras/GLM-4.6-REAP-218B-A32B
For a convenient overview and download list, visit our model page for this model.
static quants are available at https://huggingface.co/mradermacher/GLM-4.6-REAP-218B-A32B-GGUF
Usage
If you are unsure how to use GGUF files, refer to one of TheBloke's
READMEs for
more details, including on how to concatenate multi-part files.
Provided Quants
(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
| Link | Type | Size/GB | Notes |
|---|---|---|---|
| GGUF | imatrix | 0.5 | imatrix file (for creating your own quants) |
| GGUF | i1-IQ1_S | 44.4 | for the desperate |
| GGUF | i1-IQ1_M | 49.2 | mostly desperate |
| PART 1 PART 2 | i1-IQ2_XXS | 57.3 | |
| PART 1 PART 2 | i1-IQ2_XS | 64.0 | |
| PART 1 PART 2 | i1-IQ2_S | 65.2 | |
| PART 1 PART 2 | i1-IQ2_M | 71.7 | |
| PART 1 PART 2 | i1-Q2KS | 73.9 | very low quality |
| PART 1 PART 2 | i1-Q2K | 79.9 | IQ3XXS probably better |
| PART 1 PART 2 | i1-IQ3_XXS | 84.1 | lower quality |
| PART 1 PART 2 | i1-IQ3_XS | 89.2 | |
| PART 1 PART 2 | i1-Q3KS | 94.4 | IQ3_XS probably better |
| PART 1 PART 2 | i1-IQ3S | 94.5 | beats Q3K* |
| PART 1 PART 2 | i1-IQ3_M | 96.1 | |
| PART 1 PART 2 PART 3 | i1-Q3KM | 104.5 | IQ3S probably better |
| PART 1 PART 2 PART 3 | i1-Q3KL | 113.6 | IQ3M probably better |
| PART 1 PART 2 PART 3 | i1-IQ4_XS | 116.6 | |
| PART 1 PART 2 PART 3 | i1-Q4_0 | 123.7 | fast, low quality |
| PART 1 PART 2 PART 3 | i1-Q4K_S | 124.1 | optimal size/speed/quality |
| PART 1 PART 2 PART 3 | i1-Q4K_M | 131.9 | fast, recommended |
| PART 1 PART 2 PART 3 | i1-Q4_1 | 136.9 | |
| PART 1 PART 2 PART 3 PART 4 | i1-Q5KS | 150.5 | |
| PART 1 PART 2 PART 3 PART 4 | i1-Q5KM | 154.9 | |
| PART 1 PART 2 PART 3 PART 4 | i1-Q6K | 179.4 | practically like static Q6K |
And here are Artefact2's thoughts on the matter:
https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
FAQ / Model Request
See https://huggingface.co/mradermacher/model_requests for some answers to
questions you might have and/or if you want some other model quantized.
Thanks
I thank my company, nethype GmbH, for letting
me use its servers and providing upgrades to my workstation to enable
this work in my free time. Additional thanks to @nicoboss for giving me access to his private supercomputer, enabling me to provide many more imatrix quants, at much higher quality, than I would otherwise be able to.