TheBloke commited on
Commit
906e7a7
1 Parent(s): 3b2281a

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +37 -0
README.md CHANGED
@@ -115,8 +115,45 @@ Refer to the Provided Files table below to see what files use which methods, and
115
  | [nous-hermes-llama2-70b.Q4_K_M.gguf](https://huggingface.co/TheBloke/Nous-Hermes-Llama2-70B-GGUF/blob/main/nous-hermes-llama2-70b.Q4_K_M.gguf) | Q4_K_M | 4 | 41.69 GB| 44.19 GB | medium, balanced quality - recommended |
116
  | [nous-hermes-llama2-70b.Q5_K_S.gguf](https://huggingface.co/TheBloke/Nous-Hermes-Llama2-70B-GGUF/blob/main/nous-hermes-llama2-70b.Q5_K_S.gguf) | Q5_K_S | 5 | 47.74 GB| 50.24 GB | large, low quality loss - recommended |
117
  | [nous-hermes-llama2-70b.Q5_K_M.gguf](https://huggingface.co/TheBloke/Nous-Hermes-Llama2-70B-GGUF/blob/main/nous-hermes-llama2-70b.Q5_K_M.gguf) | Q5_K_M | 5 | 49.03 GB| 51.53 GB | large, very low quality loss - recommended |
 
 
118
 
119
  **Note**: the above RAM figures assume no GPU offloading. If layers are offloaded to the GPU, this will reduce RAM usage and use VRAM instead.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
120
  <!-- README_GGUF.md-provided-files end -->
121
 
122
  <!-- README_GGUF.md-how-to-run start -->
 
115
  | [nous-hermes-llama2-70b.Q4_K_M.gguf](https://huggingface.co/TheBloke/Nous-Hermes-Llama2-70B-GGUF/blob/main/nous-hermes-llama2-70b.Q4_K_M.gguf) | Q4_K_M | 4 | 41.69 GB| 44.19 GB | medium, balanced quality - recommended |
116
  | [nous-hermes-llama2-70b.Q5_K_S.gguf](https://huggingface.co/TheBloke/Nous-Hermes-Llama2-70B-GGUF/blob/main/nous-hermes-llama2-70b.Q5_K_S.gguf) | Q5_K_S | 5 | 47.74 GB| 50.24 GB | large, low quality loss - recommended |
117
  | [nous-hermes-llama2-70b.Q5_K_M.gguf](https://huggingface.co/TheBloke/Nous-Hermes-Llama2-70B-GGUF/blob/main/nous-hermes-llama2-70b.Q5_K_M.gguf) | Q5_K_M | 5 | 49.03 GB| 51.53 GB | large, very low quality loss - recommended |
118
+ | nous-hermes-llama2-70b.Q6_K.bin | q6_K | 6 | 56.82 GB | 59.32 GB | very large, extremely low quality loss |
119
+ | nous-hermes-llama2-70b.Q8_0.bin | q8_0 | 8 | 73.29 GB | 75.79 GB | very large, extremely low quality loss - not recommended |
120
 
121
  **Note**: the above RAM figures assume no GPU offloading. If layers are offloaded to the GPU, this will reduce RAM usage and use VRAM instead.
122
+
123
+ ### Q6_K and Q8_0 files are split and require joining
124
+
125
+ **Note:** HF does not support uploading files larger than 50GB. Therefore I have uploaded the Q6_K and Q8_0 files as split files.
126
+
127
+ <details>
128
+ <summary>Click for instructions regarding Q6_K and Q8_0 files</summary>
129
+
130
+ ### q6_K
131
+ Please download:
132
+ * `nous-hermes-llama2-70b.Q6_K.gguf-split-a`
133
+ * `nous-hermes-llama2-70b.Q6_K.gguf-split-b`
134
+
135
+ ### q8_0
136
+ Please download:
137
+ * `nous-hermes-llama2-70b.Q8_0.gguf-split-a`
138
+ * `nous-hermes-llama2-70b.Q8_0.gguf-split-b`
139
+
140
+ To join the files, do the following:
141
+
142
+ Linux:
143
+ ```
144
+ cat nous-hermes-llama2-70b.Q6_K.gguf-split-* > nous-hermes-llama2-70b.Q6_K.gguf && rm nous-hermes-llama2-70b.Q6_K.gguf-split-*
145
+ cat nous-hermes-llama2-70b.Q8_0.gguf-split-* > nous-hermes-llama2-70b.Q8_0.gguf && rm nous-hermes-llama2-70b.Q8_0.gguf-split-*
146
+ ```
147
+ Windows command line:
148
+ ```
149
+ COPY /B nous-hermes-llama2-70b.Q6_K.gguf-split-a + nous-hermes-llama2-70b.Q6_K.gguf-split-b nous-hermes-llama2-70b.Q6_K.gguf
150
+ COPY /B nous-hermes-llama2-70b.Q8_0.gguf-split-a + nous-hermes-llama2-70b.Q8_0.gguf-split-b nous-hermes-llama2-70b.Q8_0.gguf
151
+ del nous-hermes-llama2-70b.Q6_K.gguf-split-a nous-hermes-llama2-70b.Q6_K.gguf-split-b
152
+ del nous-hermes-llama2-70b.Q8_0.gguf-split-a nous-hermes-llama2-70b.Q8_0.gguf-split-b
153
+ ```
154
+
155
+ </details>
156
+
157
  <!-- README_GGUF.md-provided-files end -->
158
 
159
  <!-- README_GGUF.md-how-to-run start -->