mradermacher commited on
Commit
5c24aad
·
verified ·
1 Parent(s): 08b65c7

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +11 -1
README.md CHANGED
@@ -4,7 +4,8 @@ language:
4
  - en
5
  library_name: transformers
6
  license: apache-2.0
7
- no_imatrtix: "Missing importance matrix for tensor blk.29.ffn_down_exps.weight in a very low-bit quantization"
 
8
  quantized_by: mradermacher
9
  ---
10
  ## About
@@ -36,10 +37,19 @@ more details, including on how to concatenate multi-part files.
36
  |:-----|:-----|--------:|:------|
37
  | [PART 1](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-Q2_K.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-Q2_K.gguf.part2of2) | i1-Q2_K | 52.2 | IQ3_XXS probably better |
38
  | [PART 1](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-IQ3_XXS.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-IQ3_XXS.gguf.part2of2) | i1-IQ3_XXS | 55.0 | lower quality |
 
 
 
39
  | [PART 1](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-IQ3_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-IQ3_M.gguf.part2of2) | i1-IQ3_M | 64.6 | |
40
  | [PART 1](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-Q3_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-Q3_K_M.gguf.part2of2) | i1-Q3_K_M | 67.9 | IQ3_S probably better |
 
 
 
41
  | [PART 1](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-Q4_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-Q4_K_S.gguf.part2of2) | i1-Q4_K_S | 80.6 | optimal size/speed/quality |
42
  | [PART 1](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-Q4_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-Q4_K_M.gguf.part2of2) | i1-Q4_K_M | 85.7 | fast, recommended |
 
 
 
43
 
44
  Here is a handy graph by ikawrakow comparing some lower-quality quant
45
  types (lower is better):
 
4
  - en
5
  library_name: transformers
6
  license: apache-2.0
7
+ no_imatrtix: Missing importance matrix for tensor blk.29.ffn_down_exps.weight in a
8
+ very low-bit quantization
9
  quantized_by: mradermacher
10
  ---
11
  ## About
 
37
  |:-----|:-----|--------:|:------|
38
  | [PART 1](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-Q2_K.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-Q2_K.gguf.part2of2) | i1-Q2_K | 52.2 | IQ3_XXS probably better |
39
  | [PART 1](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-IQ3_XXS.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-IQ3_XXS.gguf.part2of2) | i1-IQ3_XXS | 55.0 | lower quality |
40
+ | [PART 1](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-IQ3_XS.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-IQ3_XS.gguf.part2of2) | i1-IQ3_XS | 58.3 | |
41
+ | [PART 1](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-IQ3_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-IQ3_S.gguf.part2of2) | i1-IQ3_S | 61.6 | beats Q3_K* |
42
+ | [PART 1](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-Q3_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-Q3_K_S.gguf.part2of2) | i1-Q3_K_S | 61.6 | IQ3_XS probably better |
43
  | [PART 1](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-IQ3_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-IQ3_M.gguf.part2of2) | i1-IQ3_M | 64.6 | |
44
  | [PART 1](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-Q3_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-Q3_K_M.gguf.part2of2) | i1-Q3_K_M | 67.9 | IQ3_S probably better |
45
+ | [PART 1](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-Q3_K_L.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-Q3_K_L.gguf.part2of2) | i1-Q3_K_L | 72.7 | IQ3_M probably better |
46
+ | [PART 1](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-IQ4_XS.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-IQ4_XS.gguf.part2of2) | i1-IQ4_XS | 75.6 | |
47
+ | [PART 1](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-Q4_0.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-Q4_0.gguf.part2of2) | i1-Q4_0 | 80.0 | fast, low quality |
48
  | [PART 1](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-Q4_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-Q4_K_S.gguf.part2of2) | i1-Q4_K_S | 80.6 | optimal size/speed/quality |
49
  | [PART 1](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-Q4_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-Q4_K_M.gguf.part2of2) | i1-Q4_K_M | 85.7 | fast, recommended |
50
+ | [PART 1](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-Q5_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-Q5_K_S.gguf.part2of2) | i1-Q5_K_S | 97.1 | |
51
+ | [PART 1](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-Q5_K_M.gguf.part1of3) [PART 2](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-Q5_K_M.gguf.part2of3) [PART 3](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-Q5_K_M.gguf.part3of3) | i1-Q5_K_M | 100.1 | |
52
+ | [PART 1](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-Q6_K.gguf.part1of3) [PART 2](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-Q6_K.gguf.part2of3) [PART 3](https://huggingface.co/mradermacher/Tess-3-Mixtral-8x22B-i1-GGUF/resolve/main/Tess-3-Mixtral-8x22B.i1-Q6_K.gguf.part3of3) | i1-Q6_K | 115.6 | practically like static Q6_K |
53
 
54
  Here is a handy graph by ikawrakow comparing some lower-quality quant
55
  types (lower is better):