mradermacher commited on
Commit
9ba57ed
·
verified ·
1 Parent(s): db4416d

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -2
README.md CHANGED
@@ -36,14 +36,13 @@ more details, including on how to concatenate multi-part files.
36
  | [PART 1](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.IQ4_XS.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.IQ4_XS.gguf.part2of2) | IQ4_XS | 65.2 | |
37
  | [PART 1](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.Q4_0.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.Q4_0.gguf.part2of2) | Q4_0 | 68.2 | fast, low quality |
38
  | [PART 1](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.Q4_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.Q4_K_S.gguf.part2of2) | Q4_K_S | 68.7 | fast, recommended |
39
- | [PART 1](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.IQ4_NL.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.IQ4_NL.gguf.part2of2) | IQ4_NL | 68.8 | slightly worse than Q4_K_S |
40
  | [PART 1](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.Q4_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.Q4_K_M.gguf.part2of2) | Q4_K_M | 72.6 | fast, recommended |
41
  | [PART 1](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.Q5_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.Q5_K_S.gguf.part2of2) | Q5_K_S | 83.2 | |
42
  | [PART 1](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.Q5_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.Q5_K_M.gguf.part2of2) | Q5_K_M | 85.4 | |
43
  | [PART 1](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.Q6_K.gguf.part1of3) [PART 2](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.Q6_K.gguf.part2of3) [PART 3](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.Q6_K.gguf.part3of3) | Q6_K | 99.1 | very good quality |
44
  | [PART 1](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.Q8_0.gguf.part1of3) [PART 2](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.Q8_0.gguf.part2of3) [PART 3](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.Q8_0.gguf.part3of3) | Q8_0 | 128.2 | fast, best quality |
45
 
46
-
47
  Here is a handy graph by ikawrakow comparing some lower-quality quant
48
  types (lower is better):
49
 
 
36
  | [PART 1](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.IQ4_XS.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.IQ4_XS.gguf.part2of2) | IQ4_XS | 65.2 | |
37
  | [PART 1](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.Q4_0.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.Q4_0.gguf.part2of2) | Q4_0 | 68.2 | fast, low quality |
38
  | [PART 1](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.Q4_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.Q4_K_S.gguf.part2of2) | Q4_K_S | 68.7 | fast, recommended |
39
+ | [PART 1](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.IQ4_NL.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.IQ4_NL.gguf.part2of2) | IQ4_NL | 68.8 | prefer IQ4_XS |
40
  | [PART 1](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.Q4_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.Q4_K_M.gguf.part2of2) | Q4_K_M | 72.6 | fast, recommended |
41
  | [PART 1](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.Q5_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.Q5_K_S.gguf.part2of2) | Q5_K_S | 83.2 | |
42
  | [PART 1](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.Q5_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.Q5_K_M.gguf.part2of2) | Q5_K_M | 85.4 | |
43
  | [PART 1](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.Q6_K.gguf.part1of3) [PART 2](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.Q6_K.gguf.part2of3) [PART 3](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.Q6_K.gguf.part3of3) | Q6_K | 99.1 | very good quality |
44
  | [PART 1](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.Q8_0.gguf.part1of3) [PART 2](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.Q8_0.gguf.part2of3) [PART 3](https://huggingface.co/mradermacher/Bernstein-120b-GGUF/resolve/main/Bernstein-120b.Q8_0.gguf.part3of3) | Q8_0 | 128.2 | fast, best quality |
45
 
 
46
  Here is a handy graph by ikawrakow comparing some lower-quality quant
47
  types (lower is better):
48