Transformers
GGUF
Inference Endpoints
imatrix
mradermacher commited on
Commit
28c73f7
·
verified ·
1 Parent(s): 92d6d2a

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +15 -0
README.md CHANGED
@@ -39,13 +39,28 @@ more details, including on how to concatenate multi-part files.
39
 
40
  | Link | Type | Size/GB | Notes |
41
  |:-----|:-----|--------:|:------|
 
 
 
 
 
42
  | [PART 1](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-IQ2_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-IQ2_M.gguf.part2of2) | i1-IQ2_M | 61.3 | |
 
43
  | [PART 1](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q2_K.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q2_K.gguf.part2of2) | i1-Q2_K | 66.9 | IQ3_XXS probably better |
44
  | [PART 1](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-IQ3_XXS.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-IQ3_XXS.gguf.part2of2) | i1-IQ3_XXS | 69.5 | lower quality |
 
 
 
45
  | [PART 1](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-IQ3_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-IQ3_M.gguf.part2of2) | i1-IQ3_M | 81.5 | |
46
  | [PART 1](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q3_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q3_K_M.gguf.part2of2) | i1-Q3_K_M | 85.6 | IQ3_S probably better |
 
 
47
  | [PART 1](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q4_K_S.gguf.part1of3) [PART 2](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q4_K_S.gguf.part2of3) [PART 3](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q4_K_S.gguf.part3of3) | i1-Q4_K_S | 101.6 | optimal size/speed/quality |
 
48
  | [PART 1](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q4_K_M.gguf.part1of3) [PART 2](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q4_K_M.gguf.part2of3) [PART 3](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q4_K_M.gguf.part3of3) | i1-Q4_K_M | 108.9 | fast, recommended |
 
 
 
49
  | [PART 1](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q6_K.gguf.part1of3) [PART 2](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q6_K.gguf.part2of3) [PART 3](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q6_K.gguf.part3of3) | i1-Q6_K | 147.6 | practically like static Q6_K |
50
 
51
  Here is a handy graph by ikawrakow comparing some lower-quality quant
 
39
 
40
  | Link | Type | Size/GB | Notes |
41
  |:-----|:-----|--------:|:------|
42
+ | [GGUF](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-IQ1_S.gguf) | i1-IQ1_S | 38.4 | for the desperate |
43
+ | [GGUF](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-IQ1_M.gguf) | i1-IQ1_M | 42.0 | mostly desperate |
44
+ | [GGUF](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 47.9 | |
45
+ | [PART 1](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-IQ2_XS.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-IQ2_XS.gguf.part2of2) | i1-IQ2_XS | 53.1 | |
46
+ | [PART 1](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-IQ2_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-IQ2_S.gguf.part2of2) | i1-IQ2_S | 56.6 | |
47
  | [PART 1](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-IQ2_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-IQ2_M.gguf.part2of2) | i1-IQ2_M | 61.3 | |
48
+ | [PART 1](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q2_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q2_K_S.gguf.part2of2) | i1-Q2_K_S | 61.8 | very low quality |
49
  | [PART 1](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q2_K.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q2_K.gguf.part2of2) | i1-Q2_K | 66.9 | IQ3_XXS probably better |
50
  | [PART 1](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-IQ3_XXS.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-IQ3_XXS.gguf.part2of2) | i1-IQ3_XXS | 69.5 | lower quality |
51
+ | [PART 1](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-IQ3_XS.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-IQ3_XS.gguf.part2of2) | i1-IQ3_XS | 75.4 | |
52
+ | [PART 1](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-IQ3_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-IQ3_S.gguf.part2of2) | i1-IQ3_S | 77.9 | beats Q3_K* |
53
+ | [PART 1](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q3_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q3_K_S.gguf.part2of2) | i1-Q3_K_S | 77.9 | IQ3_XS probably better |
54
  | [PART 1](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-IQ3_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-IQ3_M.gguf.part2of2) | i1-IQ3_M | 81.5 | |
55
  | [PART 1](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q3_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q3_K_M.gguf.part2of2) | i1-Q3_K_M | 85.6 | IQ3_S probably better |
56
+ | [PART 1](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q3_K_L.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q3_K_L.gguf.part2of2) | i1-Q3_K_L | 92.1 | IQ3_M probably better |
57
+ | [PART 1](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-IQ4_XS.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-IQ4_XS.gguf.part2of2) | i1-IQ4_XS | 96.0 | |
58
  | [PART 1](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q4_K_S.gguf.part1of3) [PART 2](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q4_K_S.gguf.part2of3) [PART 3](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q4_K_S.gguf.part3of3) | i1-Q4_K_S | 101.6 | optimal size/speed/quality |
59
+ | [PART 1](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q4_0.gguf.part1of3) [PART 2](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q4_0.gguf.part2of3) [PART 3](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q4_0.gguf.part3of3) | i1-Q4_0 | 102.1 | fast, low quality |
60
  | [PART 1](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q4_K_M.gguf.part1of3) [PART 2](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q4_K_M.gguf.part2of3) [PART 3](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q4_K_M.gguf.part3of3) | i1-Q4_K_M | 108.9 | fast, recommended |
61
+ | [PART 1](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q4_1.gguf.part1of3) [PART 2](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q4_1.gguf.part2of3) [PART 3](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q4_1.gguf.part3of3) | i1-Q4_1 | 112.7 | |
62
+ | [PART 1](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q5_K_S.gguf.part1of3) [PART 2](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q5_K_S.gguf.part2of3) [PART 3](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q5_K_S.gguf.part3of3) | i1-Q5_K_S | 123.9 | |
63
+ | [PART 1](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q5_K_M.gguf.part1of3) [PART 2](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q5_K_M.gguf.part2of3) [PART 3](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q5_K_M.gguf.part3of3) | i1-Q5_K_M | 131.1 | |
64
  | [PART 1](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q6_K.gguf.part1of3) [PART 2](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q6_K.gguf.part2of3) [PART 3](https://huggingface.co/mradermacher/falcon-180B-i1-GGUF/resolve/main/falcon-180B.i1-Q6_K.gguf.part3of3) | i1-Q6_K | 147.6 | practically like static Q6_K |
65
 
66
  Here is a handy graph by ikawrakow comparing some lower-quality quant