Hugging Face's logo

cmh
/
phi-4_exl3

Text Generation
English
Model card Files Community
cmh commited on
Commit
704e411
·
verified ·
1 Parent(s): ef1d52e

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +6 -6
README.md CHANGED
@@ -17,12 +17,12 @@ pipeline_tag: text-generation
17
 
18
  | | Quant type | File Size | Vram*|
19
  | -------- | ---------- | --------- | -------- |
20
- | [phi-4_3bpw](https://huggingface.co/cmh/phi-4_exl3/tree/3bpw) | 3 bits per weight | 6.08 GB | **9.4 GB** |
21
- | [phi-4_4bpw](https://huggingface.co/cmh/phi-4_exl3/tree/4bpw) | 4 bits per weight | 7.67 GB | **11.0 GB** |
22
- | [phi-4_5bpw](https://huggingface.co/cmh/phi-4_exl3/tree/5bpw) | 5 bits per weight | 9.25 GB | **12,6 GB** |
23
- | [phi-4_6bpw](https://huggingface.co/cmh/phi-4_exl3/tree/6bpw) | 6 bits per weight | 10.8 GB | **14,2 GB** |
24
- | [phi-4_7bpw](https://huggingface.co/cmh/phi-4_exl3/tree/7bpw) | 7 bits per weight | 12.4 GB | **15,8 GB** |
25
- | [phi-4_8bpw](https://huggingface.co/cmh/phi-4_exl3/tree/8bpw) | 8 bits per weight | 14.0 GB | **17,3 GB** |
26
 
27
  <sub>*approximate value at **16k context**.<sup>
28
 
 
17
 
18
  | | Quant type | File Size | Vram*|
19
  | -------- | ---------- | --------- | -------- |
20
+ | [phi-4_3bpw](https://huggingface.co/cmh/phi-4_exl3/tree/3bpw) | 3 bits per weight | 6.53 GB | **9.4 GB** |
21
+ | [phi-4_4bpw](https://huggingface.co/cmh/phi-4_exl3/tree/4bpw) | 4 bits per weight | 8.24 GB | **11.0 GB** |
22
+ | [phi-4_5bpw](https://huggingface.co/cmh/phi-4_exl3/tree/5bpw) | 5 bits per weight | 9.94 GB | **12,6 GB** |
23
+ | [phi-4_6bpw](https://huggingface.co/cmh/phi-4_exl3/tree/6bpw) | 6 bits per weight | 11.65 GB | **14,2 GB** |
24
+ | [phi-4_7bpw](https://huggingface.co/cmh/phi-4_exl3/tree/7bpw) | 7 bits per weight | 13.35 GB | **15,8 GB** |
25
+ | [phi-4_8bpw](https://huggingface.co/cmh/phi-4_exl3/tree/8bpw) | 8 bits per weight | 15.05 GB | **17,3 GB** |
26
 
27
  <sub>*approximate value at **16k context**.<sup>
28