Hugging Face's logo

mradermacher
/
Geneva-12B-GCv2-500k-i1-GGUF

Transformers
GGUF
text-generation-inference
unsloth
trl
gammacorpus
geneva
chat
mistral
conversational
Inference Endpoints
imatrix
Model card Files Community
mradermacher commited on
Commit
f03951c
·
verified ·
1 Parent(s): c086aae

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +5 -0
README.md CHANGED
@@ -58,15 +58,20 @@ more details, including on how to concatenate multi-part files.
58
  | [GGUF](https://huggingface.co/mradermacher/Geneva-12B-GCv2-500k-i1-GGUF/resolve/main/Geneva-12B-GCv2-500k.i1-Q2_K_S.gguf) | i1-Q2_K_S | 4.6 | very low quality |
59
  | [GGUF](https://huggingface.co/mradermacher/Geneva-12B-GCv2-500k-i1-GGUF/resolve/main/Geneva-12B-GCv2-500k.i1-Q2_K.gguf) | i1-Q2_K | 4.9 | IQ3_XXS probably better |
60
  | [GGUF](https://huggingface.co/mradermacher/Geneva-12B-GCv2-500k-i1-GGUF/resolve/main/Geneva-12B-GCv2-500k.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 5.0 | lower quality |
 
61
  | [GGUF](https://huggingface.co/mradermacher/Geneva-12B-GCv2-500k-i1-GGUF/resolve/main/Geneva-12B-GCv2-500k.i1-Q3_K_S.gguf) | i1-Q3_K_S | 5.6 | IQ3_XS probably better |
 
62
  | [GGUF](https://huggingface.co/mradermacher/Geneva-12B-GCv2-500k-i1-GGUF/resolve/main/Geneva-12B-GCv2-500k.i1-IQ3_M.gguf) | i1-IQ3_M | 5.8 | |
63
  | [GGUF](https://huggingface.co/mradermacher/Geneva-12B-GCv2-500k-i1-GGUF/resolve/main/Geneva-12B-GCv2-500k.i1-Q3_K_M.gguf) | i1-Q3_K_M | 6.2 | IQ3_S probably better |
64
  | [GGUF](https://huggingface.co/mradermacher/Geneva-12B-GCv2-500k-i1-GGUF/resolve/main/Geneva-12B-GCv2-500k.i1-Q3_K_L.gguf) | i1-Q3_K_L | 6.7 | IQ3_M probably better |
65
  | [GGUF](https://huggingface.co/mradermacher/Geneva-12B-GCv2-500k-i1-GGUF/resolve/main/Geneva-12B-GCv2-500k.i1-IQ4_XS.gguf) | i1-IQ4_XS | 6.8 | |
 
66
  | [GGUF](https://huggingface.co/mradermacher/Geneva-12B-GCv2-500k-i1-GGUF/resolve/main/Geneva-12B-GCv2-500k.i1-IQ4_NL.gguf) | i1-IQ4_NL | 7.2 | prefer IQ4_XS |
67
  | [GGUF](https://huggingface.co/mradermacher/Geneva-12B-GCv2-500k-i1-GGUF/resolve/main/Geneva-12B-GCv2-500k.i1-Q4_K_S.gguf) | i1-Q4_K_S | 7.2 | optimal size/speed/quality |
68
  | [GGUF](https://huggingface.co/mradermacher/Geneva-12B-GCv2-500k-i1-GGUF/resolve/main/Geneva-12B-GCv2-500k.i1-Q4_K_M.gguf) | i1-Q4_K_M | 7.6 | fast, recommended |
 
69
  | [GGUF](https://huggingface.co/mradermacher/Geneva-12B-GCv2-500k-i1-GGUF/resolve/main/Geneva-12B-GCv2-500k.i1-Q5_K_S.gguf) | i1-Q5_K_S | 8.6 | |
 
70
  | [GGUF](https://huggingface.co/mradermacher/Geneva-12B-GCv2-500k-i1-GGUF/resolve/main/Geneva-12B-GCv2-500k.i1-Q6_K.gguf) | i1-Q6_K | 10.2 | practically like static Q6_K |
71
 
72
  Here is a handy graph by ikawrakow comparing some lower-quality quant
 
58
  | [GGUF](https://huggingface.co/mradermacher/Geneva-12B-GCv2-500k-i1-GGUF/resolve/main/Geneva-12B-GCv2-500k.i1-Q2_K_S.gguf) | i1-Q2_K_S | 4.6 | very low quality |
59
  | [GGUF](https://huggingface.co/mradermacher/Geneva-12B-GCv2-500k-i1-GGUF/resolve/main/Geneva-12B-GCv2-500k.i1-Q2_K.gguf) | i1-Q2_K | 4.9 | IQ3_XXS probably better |
60
  | [GGUF](https://huggingface.co/mradermacher/Geneva-12B-GCv2-500k-i1-GGUF/resolve/main/Geneva-12B-GCv2-500k.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 5.0 | lower quality |
61
+ | [GGUF](https://huggingface.co/mradermacher/Geneva-12B-GCv2-500k-i1-GGUF/resolve/main/Geneva-12B-GCv2-500k.i1-IQ3_XS.gguf) | i1-IQ3_XS | 5.4 | |
62
  | [GGUF](https://huggingface.co/mradermacher/Geneva-12B-GCv2-500k-i1-GGUF/resolve/main/Geneva-12B-GCv2-500k.i1-Q3_K_S.gguf) | i1-Q3_K_S | 5.6 | IQ3_XS probably better |
63
+ | [GGUF](https://huggingface.co/mradermacher/Geneva-12B-GCv2-500k-i1-GGUF/resolve/main/Geneva-12B-GCv2-500k.i1-IQ3_S.gguf) | i1-IQ3_S | 5.7 | beats Q3_K* |
64
  | [GGUF](https://huggingface.co/mradermacher/Geneva-12B-GCv2-500k-i1-GGUF/resolve/main/Geneva-12B-GCv2-500k.i1-IQ3_M.gguf) | i1-IQ3_M | 5.8 | |
65
  | [GGUF](https://huggingface.co/mradermacher/Geneva-12B-GCv2-500k-i1-GGUF/resolve/main/Geneva-12B-GCv2-500k.i1-Q3_K_M.gguf) | i1-Q3_K_M | 6.2 | IQ3_S probably better |
66
  | [GGUF](https://huggingface.co/mradermacher/Geneva-12B-GCv2-500k-i1-GGUF/resolve/main/Geneva-12B-GCv2-500k.i1-Q3_K_L.gguf) | i1-Q3_K_L | 6.7 | IQ3_M probably better |
67
  | [GGUF](https://huggingface.co/mradermacher/Geneva-12B-GCv2-500k-i1-GGUF/resolve/main/Geneva-12B-GCv2-500k.i1-IQ4_XS.gguf) | i1-IQ4_XS | 6.8 | |
68
+ | [GGUF](https://huggingface.co/mradermacher/Geneva-12B-GCv2-500k-i1-GGUF/resolve/main/Geneva-12B-GCv2-500k.i1-Q4_0.gguf) | i1-Q4_0 | 7.2 | fast, low quality |
69
  | [GGUF](https://huggingface.co/mradermacher/Geneva-12B-GCv2-500k-i1-GGUF/resolve/main/Geneva-12B-GCv2-500k.i1-IQ4_NL.gguf) | i1-IQ4_NL | 7.2 | prefer IQ4_XS |
70
  | [GGUF](https://huggingface.co/mradermacher/Geneva-12B-GCv2-500k-i1-GGUF/resolve/main/Geneva-12B-GCv2-500k.i1-Q4_K_S.gguf) | i1-Q4_K_S | 7.2 | optimal size/speed/quality |
71
  | [GGUF](https://huggingface.co/mradermacher/Geneva-12B-GCv2-500k-i1-GGUF/resolve/main/Geneva-12B-GCv2-500k.i1-Q4_K_M.gguf) | i1-Q4_K_M | 7.6 | fast, recommended |
72
+ | [GGUF](https://huggingface.co/mradermacher/Geneva-12B-GCv2-500k-i1-GGUF/resolve/main/Geneva-12B-GCv2-500k.i1-Q4_1.gguf) | i1-Q4_1 | 7.9 | |
73
  | [GGUF](https://huggingface.co/mradermacher/Geneva-12B-GCv2-500k-i1-GGUF/resolve/main/Geneva-12B-GCv2-500k.i1-Q5_K_S.gguf) | i1-Q5_K_S | 8.6 | |
74
+ | [GGUF](https://huggingface.co/mradermacher/Geneva-12B-GCv2-500k-i1-GGUF/resolve/main/Geneva-12B-GCv2-500k.i1-Q5_K_M.gguf) | i1-Q5_K_M | 8.8 | |
75
  | [GGUF](https://huggingface.co/mradermacher/Geneva-12B-GCv2-500k-i1-GGUF/resolve/main/Geneva-12B-GCv2-500k.i1-Q6_K.gguf) | i1-Q6_K | 10.2 | practically like static Q6_K |
76
 
77
  Here is a handy graph by ikawrakow comparing some lower-quality quant