Hugging Face's logo

mradermacher
/
Qwen2.5-72B-GGUF

Transformers
GGUF
English
Inference Endpoints
conversational
Model card Files Community
mradermacher commited on
Commit
169bae2
•
1 Parent(s): d1bfdde

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +14 -1
README.md CHANGED
@@ -18,7 +18,7 @@ quantized_by: mradermacher
18
  static quants of https://huggingface.co/Qwen/Qwen2.5-72B
19
 
20
  <!-- provided-files -->
21
- weighted/imatrix quants seem not to be available (by me) at this time. If they do not show up a week or so after the static ones, I have probably not planned for them. Feel free to request them by opening a Community Discussion.
22
  ## Usage
23
 
24
  If you are unsure how to use GGUF files, refer to one of [TheBloke's
@@ -31,7 +31,20 @@ more details, including on how to concatenate multi-part files.
31
 
32
  | Link | Type | Size/GB | Notes |
33
  |:-----|:-----|--------:|:------|
 
 
34
  | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-GGUF/resolve/main/Qwen2.5-72B.IQ3_S.gguf) | IQ3_S | 34.6 | beats Q3_K* |
 
 
 
 
 
 
 
 
 
 
 
35
 
36
  Here is a handy graph by ikawrakow comparing some lower-quality quant
37
  types (lower is better):
 
18
  static quants of https://huggingface.co/Qwen/Qwen2.5-72B
19
 
20
  <!-- provided-files -->
21
+ weighted/imatrix quants are available at https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF
22
  ## Usage
23
 
24
  If you are unsure how to use GGUF files, refer to one of [TheBloke's
 
31
 
32
  | Link | Type | Size/GB | Notes |
33
  |:-----|:-----|--------:|:------|
34
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-GGUF/resolve/main/Qwen2.5-72B.Q2_K.gguf) | Q2_K | 29.9 | |
35
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-GGUF/resolve/main/Qwen2.5-72B.IQ3_XS.gguf) | IQ3_XS | 32.9 | |
36
  | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-GGUF/resolve/main/Qwen2.5-72B.IQ3_S.gguf) | IQ3_S | 34.6 | beats Q3_K* |
37
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-GGUF/resolve/main/Qwen2.5-72B.Q3_K_S.gguf) | Q3_K_S | 34.6 | |
38
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-GGUF/resolve/main/Qwen2.5-72B.IQ3_M.gguf) | IQ3_M | 35.6 | |
39
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-GGUF/resolve/main/Qwen2.5-72B.Q3_K_M.gguf) | Q3_K_M | 37.8 | lower quality |
40
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-GGUF/resolve/main/Qwen2.5-72B.Q3_K_L.gguf) | Q3_K_L | 39.6 | |
41
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-GGUF/resolve/main/Qwen2.5-72B.IQ4_XS.gguf) | IQ4_XS | 40.3 | |
42
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-GGUF/resolve/main/Qwen2.5-72B.Q4_K_S.gguf) | Q4_K_S | 44.0 | fast, recommended |
43
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-GGUF/resolve/main/Qwen2.5-72B.Q4_K_M.gguf) | Q4_K_M | 47.5 | fast, recommended |
44
+ | [PART 1](https://huggingface.co/mradermacher/Qwen2.5-72B-GGUF/resolve/main/Qwen2.5-72B.Q5_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Qwen2.5-72B-GGUF/resolve/main/Qwen2.5-72B.Q5_K_S.gguf.part2of2) | Q5_K_S | 51.5 | |
45
+ | [PART 1](https://huggingface.co/mradermacher/Qwen2.5-72B-GGUF/resolve/main/Qwen2.5-72B.Q5_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Qwen2.5-72B-GGUF/resolve/main/Qwen2.5-72B.Q5_K_M.gguf.part2of2) | Q5_K_M | 54.5 | |
46
+ | [PART 1](https://huggingface.co/mradermacher/Qwen2.5-72B-GGUF/resolve/main/Qwen2.5-72B.Q6_K.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Qwen2.5-72B-GGUF/resolve/main/Qwen2.5-72B.Q6_K.gguf.part2of2) | Q6_K | 64.4 | very good quality |
47
+ | [PART 1](https://huggingface.co/mradermacher/Qwen2.5-72B-GGUF/resolve/main/Qwen2.5-72B.Q8_0.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Qwen2.5-72B-GGUF/resolve/main/Qwen2.5-72B.Q8_0.gguf.part2of2) | Q8_0 | 77.4 | fast, best quality |
48
 
49
  Here is a handy graph by ikawrakow comparing some lower-quality quant
50
  types (lower is better):