Hugging Face's logo

mradermacher
/
Hermes-4-70B-i1-GGUF

Transformers
GGUF
English
Llama-3.1
instruct
finetune
reasoning
hybrid-mode
chatml
function calling
tool use
json mode
structured outputs
atropos
dataforge
long context
roleplaying
chat
imatrix
conversational
Model card Files
xet
Community
mradermacher commited on
Commit
78b99ab
·
verified ·
1 Parent(s): fd3d22e

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +23 -0
README.md CHANGED
@@ -54,6 +54,29 @@ more details, including on how to concatenate multi-part files.
54
  | Link | Type | Size/GB | Notes |
55
  |:-----|:-----|--------:|:------|
56
  | [GGUF](https://huggingface.co/mradermacher/Hermes-4-70B-i1-GGUF/resolve/main/Hermes-4-70B.imatrix.gguf) | imatrix | 0.1 | imatrix file (for creating your own qwuants) |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
57
 
58
  Here is a handy graph by ikawrakow comparing some lower-quality quant
59
  types (lower is better):
 
54
  | Link | Type | Size/GB | Notes |
55
  |:-----|:-----|--------:|:------|
56
  | [GGUF](https://huggingface.co/mradermacher/Hermes-4-70B-i1-GGUF/resolve/main/Hermes-4-70B.imatrix.gguf) | imatrix | 0.1 | imatrix file (for creating your own qwuants) |
57
+ | [GGUF](https://huggingface.co/mradermacher/Hermes-4-70B-i1-GGUF/resolve/main/Hermes-4-70B.i1-IQ1_S.gguf) | i1-IQ1_S | 15.4 | for the desperate |
58
+ | [GGUF](https://huggingface.co/mradermacher/Hermes-4-70B-i1-GGUF/resolve/main/Hermes-4-70B.i1-IQ1_M.gguf) | i1-IQ1_M | 16.9 | mostly desperate |
59
+ | [GGUF](https://huggingface.co/mradermacher/Hermes-4-70B-i1-GGUF/resolve/main/Hermes-4-70B.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 19.2 | |
60
+ | [GGUF](https://huggingface.co/mradermacher/Hermes-4-70B-i1-GGUF/resolve/main/Hermes-4-70B.i1-IQ2_XS.gguf) | i1-IQ2_XS | 21.2 | |
61
+ | [GGUF](https://huggingface.co/mradermacher/Hermes-4-70B-i1-GGUF/resolve/main/Hermes-4-70B.i1-IQ2_S.gguf) | i1-IQ2_S | 22.3 | |
62
+ | [GGUF](https://huggingface.co/mradermacher/Hermes-4-70B-i1-GGUF/resolve/main/Hermes-4-70B.i1-IQ2_M.gguf) | i1-IQ2_M | 24.2 | |
63
+ | [GGUF](https://huggingface.co/mradermacher/Hermes-4-70B-i1-GGUF/resolve/main/Hermes-4-70B.i1-Q2_K_S.gguf) | i1-Q2_K_S | 24.6 | very low quality |
64
+ | [GGUF](https://huggingface.co/mradermacher/Hermes-4-70B-i1-GGUF/resolve/main/Hermes-4-70B.i1-Q2_K.gguf) | i1-Q2_K | 26.5 | IQ3_XXS probably better |
65
+ | [GGUF](https://huggingface.co/mradermacher/Hermes-4-70B-i1-GGUF/resolve/main/Hermes-4-70B.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 27.6 | lower quality |
66
+ | [GGUF](https://huggingface.co/mradermacher/Hermes-4-70B-i1-GGUF/resolve/main/Hermes-4-70B.i1-IQ3_XS.gguf) | i1-IQ3_XS | 29.4 | |
67
+ | [GGUF](https://huggingface.co/mradermacher/Hermes-4-70B-i1-GGUF/resolve/main/Hermes-4-70B.i1-IQ3_S.gguf) | i1-IQ3_S | 31.0 | beats Q3_K* |
68
+ | [GGUF](https://huggingface.co/mradermacher/Hermes-4-70B-i1-GGUF/resolve/main/Hermes-4-70B.i1-Q3_K_S.gguf) | i1-Q3_K_S | 31.0 | IQ3_XS probably better |
69
+ | [GGUF](https://huggingface.co/mradermacher/Hermes-4-70B-i1-GGUF/resolve/main/Hermes-4-70B.i1-IQ3_M.gguf) | i1-IQ3_M | 32.0 | |
70
+ | [GGUF](https://huggingface.co/mradermacher/Hermes-4-70B-i1-GGUF/resolve/main/Hermes-4-70B.i1-Q3_K_M.gguf) | i1-Q3_K_M | 34.4 | IQ3_S probably better |
71
+ | [GGUF](https://huggingface.co/mradermacher/Hermes-4-70B-i1-GGUF/resolve/main/Hermes-4-70B.i1-Q3_K_L.gguf) | i1-Q3_K_L | 37.2 | IQ3_M probably better |
72
+ | [GGUF](https://huggingface.co/mradermacher/Hermes-4-70B-i1-GGUF/resolve/main/Hermes-4-70B.i1-IQ4_XS.gguf) | i1-IQ4_XS | 38.0 | |
73
+ | [GGUF](https://huggingface.co/mradermacher/Hermes-4-70B-i1-GGUF/resolve/main/Hermes-4-70B.i1-Q4_0.gguf) | i1-Q4_0 | 40.2 | fast, low quality |
74
+ | [GGUF](https://huggingface.co/mradermacher/Hermes-4-70B-i1-GGUF/resolve/main/Hermes-4-70B.i1-Q4_K_S.gguf) | i1-Q4_K_S | 40.4 | optimal size/speed/quality |
75
+ | [GGUF](https://huggingface.co/mradermacher/Hermes-4-70B-i1-GGUF/resolve/main/Hermes-4-70B.i1-Q4_K_M.gguf) | i1-Q4_K_M | 42.6 | fast, recommended |
76
+ | [GGUF](https://huggingface.co/mradermacher/Hermes-4-70B-i1-GGUF/resolve/main/Hermes-4-70B.i1-Q4_1.gguf) | i1-Q4_1 | 44.4 | |
77
+ | [GGUF](https://huggingface.co/mradermacher/Hermes-4-70B-i1-GGUF/resolve/main/Hermes-4-70B.i1-Q5_K_S.gguf) | i1-Q5_K_S | 48.8 | |
78
+ | [GGUF](https://huggingface.co/mradermacher/Hermes-4-70B-i1-GGUF/resolve/main/Hermes-4-70B.i1-Q5_K_M.gguf) | i1-Q5_K_M | 50.0 | |
79
+ | [PART 1](https://huggingface.co/mradermacher/Hermes-4-70B-i1-GGUF/resolve/main/Hermes-4-70B.i1-Q6_K.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Hermes-4-70B-i1-GGUF/resolve/main/Hermes-4-70B.i1-Q6_K.gguf.part2of2) | i1-Q6_K | 58.0 | practically like static Q6_K |
80
 
81
  Here is a handy graph by ikawrakow comparing some lower-quality quant
82
  types (lower is better):