TheBloke commited on
Commit
efb0778
1 Parent(s): 22662e2

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +39 -4
README.md CHANGED
@@ -110,20 +110,55 @@ Refer to the Provided Files table below to see what files use which methods, and
110
 
111
  | Name | Quant method | Bits | Size | Max RAM required | Use case |
112
  | ---- | ---- | ---- | ---- | ---- | ----- |
113
- | [nous-puffin-70b.Q6_K.gguf-split-b](https://huggingface.co/TheBloke/Nous-Puffin-70B-GGUF/blob/main/nous-puffin-70b.Q6_K.gguf-split-b) | Q6_K | 6 | 19.89 GB| 22.39 GB | very large, extremely low quality loss |
114
  | [nous-puffin-70b.Q2_K.gguf](https://huggingface.co/TheBloke/Nous-Puffin-70B-GGUF/blob/main/nous-puffin-70b.Q2_K.gguf) | Q2_K | 2 | 29.11 GB| 31.61 GB | smallest, significant quality loss - not recommended for most purposes |
115
  | [nous-puffin-70b.Q3_K_S.gguf](https://huggingface.co/TheBloke/Nous-Puffin-70B-GGUF/blob/main/nous-puffin-70b.Q3_K_S.gguf) | Q3_K_S | 3 | 29.75 GB| 32.25 GB | very small, high quality loss |
116
  | [nous-puffin-70b.Q3_K_M.gguf](https://huggingface.co/TheBloke/Nous-Puffin-70B-GGUF/blob/main/nous-puffin-70b.Q3_K_M.gguf) | Q3_K_M | 3 | 33.10 GB| 35.60 GB | very small, high quality loss |
117
  | [nous-puffin-70b.Q3_K_L.gguf](https://huggingface.co/TheBloke/Nous-Puffin-70B-GGUF/blob/main/nous-puffin-70b.Q3_K_L.gguf) | Q3_K_L | 3 | 36.15 GB| 38.65 GB | small, substantial quality loss |
118
- | [nous-puffin-70b.Q8_0.gguf-split-b](https://huggingface.co/TheBloke/Nous-Puffin-70B-GGUF/blob/main/nous-puffin-70b.Q8_0.gguf-split-b) | Q8_0 | 8 | 36.53 GB| 39.03 GB | very large, extremely low quality loss - not recommended |
119
- | [nous-puffin-70b.Q6_K.gguf-split-a](https://huggingface.co/TheBloke/Nous-Puffin-70B-GGUF/blob/main/nous-puffin-70b.Q6_K.gguf-split-a) | Q6_K | 6 | 36.70 GB| 39.20 GB | very large, extremely low quality loss |
120
- | [nous-puffin-70b.Q8_0.gguf-split-a](https://huggingface.co/TheBloke/Nous-Puffin-70B-GGUF/blob/main/nous-puffin-70b.Q8_0.gguf-split-a) | Q8_0 | 8 | 36.70 GB| 39.20 GB | very large, extremely low quality loss - not recommended |
121
  | [nous-puffin-70b.Q4_K_S.gguf](https://huggingface.co/TheBloke/Nous-Puffin-70B-GGUF/blob/main/nous-puffin-70b.Q4_K_S.gguf) | Q4_K_S | 4 | 38.99 GB| 41.49 GB | small, greater quality loss |
122
  | [nous-puffin-70b.Q4_K_M.gguf](https://huggingface.co/TheBloke/Nous-Puffin-70B-GGUF/blob/main/nous-puffin-70b.Q4_K_M.gguf) | Q4_K_M | 4 | 41.38 GB| 43.88 GB | medium, balanced quality - recommended |
123
  | [nous-puffin-70b.Q5_K_S.gguf](https://huggingface.co/TheBloke/Nous-Puffin-70B-GGUF/blob/main/nous-puffin-70b.Q5_K_S.gguf) | Q5_K_S | 5 | 47.46 GB| 49.96 GB | large, low quality loss - recommended |
124
  | [nous-puffin-70b.Q5_K_M.gguf](https://huggingface.co/TheBloke/Nous-Puffin-70B-GGUF/blob/main/nous-puffin-70b.Q5_K_M.gguf) | Q5_K_M | 5 | 48.75 GB| 51.25 GB | large, very low quality loss - recommended |
 
 
125
 
126
  **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.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
127
  <!-- README_GGUF.md-provided-files end -->
128
 
129
  <!-- README_GGUF.md-how-to-run start -->
 
110
 
111
  | Name | Quant method | Bits | Size | Max RAM required | Use case |
112
  | ---- | ---- | ---- | ---- | ---- | ----- |
 
113
  | [nous-puffin-70b.Q2_K.gguf](https://huggingface.co/TheBloke/Nous-Puffin-70B-GGUF/blob/main/nous-puffin-70b.Q2_K.gguf) | Q2_K | 2 | 29.11 GB| 31.61 GB | smallest, significant quality loss - not recommended for most purposes |
114
  | [nous-puffin-70b.Q3_K_S.gguf](https://huggingface.co/TheBloke/Nous-Puffin-70B-GGUF/blob/main/nous-puffin-70b.Q3_K_S.gguf) | Q3_K_S | 3 | 29.75 GB| 32.25 GB | very small, high quality loss |
115
  | [nous-puffin-70b.Q3_K_M.gguf](https://huggingface.co/TheBloke/Nous-Puffin-70B-GGUF/blob/main/nous-puffin-70b.Q3_K_M.gguf) | Q3_K_M | 3 | 33.10 GB| 35.60 GB | very small, high quality loss |
116
  | [nous-puffin-70b.Q3_K_L.gguf](https://huggingface.co/TheBloke/Nous-Puffin-70B-GGUF/blob/main/nous-puffin-70b.Q3_K_L.gguf) | Q3_K_L | 3 | 36.15 GB| 38.65 GB | small, substantial quality loss |
 
 
 
117
  | [nous-puffin-70b.Q4_K_S.gguf](https://huggingface.co/TheBloke/Nous-Puffin-70B-GGUF/blob/main/nous-puffin-70b.Q4_K_S.gguf) | Q4_K_S | 4 | 38.99 GB| 41.49 GB | small, greater quality loss |
118
  | [nous-puffin-70b.Q4_K_M.gguf](https://huggingface.co/TheBloke/Nous-Puffin-70B-GGUF/blob/main/nous-puffin-70b.Q4_K_M.gguf) | Q4_K_M | 4 | 41.38 GB| 43.88 GB | medium, balanced quality - recommended |
119
  | [nous-puffin-70b.Q5_K_S.gguf](https://huggingface.co/TheBloke/Nous-Puffin-70B-GGUF/blob/main/nous-puffin-70b.Q5_K_S.gguf) | Q5_K_S | 5 | 47.46 GB| 49.96 GB | large, low quality loss - recommended |
120
  | [nous-puffin-70b.Q5_K_M.gguf](https://huggingface.co/TheBloke/Nous-Puffin-70B-GGUF/blob/main/nous-puffin-70b.Q5_K_M.gguf) | Q5_K_M | 5 | 48.75 GB| 51.25 GB | large, very low quality loss - recommended |
121
+ | nous-puffin-70b.Q6_K.bin | q6_K | 6 | 56.82 GB | 59.32 GB | very large, extremely low quality loss |
122
+ | nous-puffin-70b.Q8_0.bin | q8_0 | 8 | 73.29 GB | 75.79 GB | very large, extremely low quality loss - not recommended |
123
 
124
  **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.
125
+
126
+ ### Q6_K and Q8_0 files are split and require joining
127
+
128
+ **Note:** HF does not support uploading files larger than 50GB. Therefore I have uploaded the Q6_K and Q8_0 files as split files.
129
+
130
+ <details>
131
+ <summary>Click for instructions regarding Q6_K and Q8_0 files</summary>
132
+
133
+ ### q6_K
134
+ Please download:
135
+ * `nous-puffin-70b.Q6_K.gguf-split-a`
136
+ * `nous-puffin-70b.Q6_K.gguf-split-b`
137
+
138
+ ### q8_0
139
+ Please download:
140
+ * `nous-puffin-70b.Q8_0.gguf-split-a`
141
+ * `nous-puffin-70b.Q8_0.gguf-split-b`
142
+
143
+ To join the files, do the following:
144
+
145
+ Linux and macOS:
146
+ ```
147
+ cat nous-puffin-70b.Q6_K.gguf-split-* > nous-puffin-70b.Q6_K.gguf && rm nous-puffin-70b.Q6_K.gguf-split-*
148
+ cat nous-puffin-70b.Q8_0.gguf-split-* > nous-puffin-70b.Q8_0.gguf && rm nous-puffin-70b.Q8_0.gguf-split-*
149
+ ```
150
+ Windows command line:
151
+ ```
152
+ COPY /B nous-puffin-70b.Q6_K.gguf-split-a + nous-puffin-70b.Q6_K.gguf-split-b nous-puffin-70b.Q6_K.gguf
153
+ del nous-puffin-70b.Q6_K.gguf-split-a nous-puffin-70b.Q6_K.gguf-split-b
154
+
155
+ COPY /B nous-puffin-70b.Q8_0.gguf-split-a + nous-puffin-70b.Q8_0.gguf-split-b nous-puffin-70b.Q8_0.gguf
156
+ del nous-puffin-70b.Q8_0.gguf-split-a nous-puffin-70b.Q8_0.gguf-split-b
157
+ ```
158
+
159
+ </details>
160
+
161
+
162
  <!-- README_GGUF.md-provided-files end -->
163
 
164
  <!-- README_GGUF.md-how-to-run start -->