Hugging Face's logo

tensorblock
/
Llama-3-8B-Instruct-v0.9-GGUF

Text Generation
Transformers
GGUF
PyTorch
English
axolotl
finetune
facebook
meta
llama
llama-3
TensorBlock
GGUF
Eval Results
conversational
Model card Files Community
morriszms commited on
Commit
9ea9611
·
verified ·
1 Parent(s): f143760

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +20 -12
README.md CHANGED
@@ -142,8 +142,16 @@ This repo contains GGUF format model files for [MaziyarPanahi/Llama-3-8B-Instruc
142
 
143
  The files were quantized using machines provided by [TensorBlock](https://tensorblock.co/), and they are compatible with llama.cpp as of [commit b4011](https://github.com/ggerganov/llama.cpp/commit/a6744e43e80f4be6398fc7733a01642c846dce1d).
144
 
 
 
 
 
 
 
 
145
  ## Prompt template
146
 
 
147
  ```
148
  <|begin_of_text|><|start_header_id|>system<|end_header_id|>
149
 
@@ -156,18 +164,18 @@ The files were quantized using machines provided by [TensorBlock](https://tensor
156
 
157
  | Filename | Quant type | File Size | Description |
158
  | -------- | ---------- | --------- | ----------- |
159
- | [Llama-3-8B-Instruct-v0.9-Q2_K.gguf](https://huggingface.co/tensorblock/Llama-3-8B-Instruct-v0.9-GGUF/tree/main/Llama-3-8B-Instruct-v0.9-Q2_K.gguf) | Q2_K | 2.961 GB | smallest, significant quality loss - not recommended for most purposes |
160
- | [Llama-3-8B-Instruct-v0.9-Q3_K_S.gguf](https://huggingface.co/tensorblock/Llama-3-8B-Instruct-v0.9-GGUF/tree/main/Llama-3-8B-Instruct-v0.9-Q3_K_S.gguf) | Q3_K_S | 3.413 GB | very small, high quality loss |
161
- | [Llama-3-8B-Instruct-v0.9-Q3_K_M.gguf](https://huggingface.co/tensorblock/Llama-3-8B-Instruct-v0.9-GGUF/tree/main/Llama-3-8B-Instruct-v0.9-Q3_K_M.gguf) | Q3_K_M | 3.743 GB | very small, high quality loss |
162
- | [Llama-3-8B-Instruct-v0.9-Q3_K_L.gguf](https://huggingface.co/tensorblock/Llama-3-8B-Instruct-v0.9-GGUF/tree/main/Llama-3-8B-Instruct-v0.9-Q3_K_L.gguf) | Q3_K_L | 4.025 GB | small, substantial quality loss |
163
- | [Llama-3-8B-Instruct-v0.9-Q4_0.gguf](https://huggingface.co/tensorblock/Llama-3-8B-Instruct-v0.9-GGUF/tree/main/Llama-3-8B-Instruct-v0.9-Q4_0.gguf) | Q4_0 | 4.341 GB | legacy; small, very high quality loss - prefer using Q3_K_M |
164
- | [Llama-3-8B-Instruct-v0.9-Q4_K_S.gguf](https://huggingface.co/tensorblock/Llama-3-8B-Instruct-v0.9-GGUF/tree/main/Llama-3-8B-Instruct-v0.9-Q4_K_S.gguf) | Q4_K_S | 4.370 GB | small, greater quality loss |
165
- | [Llama-3-8B-Instruct-v0.9-Q4_K_M.gguf](https://huggingface.co/tensorblock/Llama-3-8B-Instruct-v0.9-GGUF/tree/main/Llama-3-8B-Instruct-v0.9-Q4_K_M.gguf) | Q4_K_M | 4.583 GB | medium, balanced quality - recommended |
166
- | [Llama-3-8B-Instruct-v0.9-Q5_0.gguf](https://huggingface.co/tensorblock/Llama-3-8B-Instruct-v0.9-GGUF/tree/main/Llama-3-8B-Instruct-v0.9-Q5_0.gguf) | Q5_0 | 5.215 GB | legacy; medium, balanced quality - prefer using Q4_K_M |
167
- | [Llama-3-8B-Instruct-v0.9-Q5_K_S.gguf](https://huggingface.co/tensorblock/Llama-3-8B-Instruct-v0.9-GGUF/tree/main/Llama-3-8B-Instruct-v0.9-Q5_K_S.gguf) | Q5_K_S | 5.215 GB | large, low quality loss - recommended |
168
- | [Llama-3-8B-Instruct-v0.9-Q5_K_M.gguf](https://huggingface.co/tensorblock/Llama-3-8B-Instruct-v0.9-GGUF/tree/main/Llama-3-8B-Instruct-v0.9-Q5_K_M.gguf) | Q5_K_M | 5.339 GB | large, very low quality loss - recommended |
169
- | [Llama-3-8B-Instruct-v0.9-Q6_K.gguf](https://huggingface.co/tensorblock/Llama-3-8B-Instruct-v0.9-GGUF/tree/main/Llama-3-8B-Instruct-v0.9-Q6_K.gguf) | Q6_K | 6.143 GB | very large, extremely low quality loss |
170
- | [Llama-3-8B-Instruct-v0.9-Q8_0.gguf](https://huggingface.co/tensorblock/Llama-3-8B-Instruct-v0.9-GGUF/tree/main/Llama-3-8B-Instruct-v0.9-Q8_0.gguf) | Q8_0 | 7.954 GB | very large, extremely low quality loss - not recommended |
171
 
172
 
173
  ## Downloading instruction
 
142
 
143
  The files were quantized using machines provided by [TensorBlock](https://tensorblock.co/), and they are compatible with llama.cpp as of [commit b4011](https://github.com/ggerganov/llama.cpp/commit/a6744e43e80f4be6398fc7733a01642c846dce1d).
144
 
145
+
146
+ <div style="text-align: left; margin: 20px 0;">
147
+ <a href="https://tensorblock.co/waitlist/client" style="display: inline-block; padding: 10px 20px; background-color: #007bff; color: white; text-decoration: none; border-radius: 5px; font-weight: bold;">
148
+ Run them on the TensorBlock client using your local machine ↗
149
+ </a>
150
+ </div>
151
+
152
  ## Prompt template
153
 
154
+
155
  ```
156
  <|begin_of_text|><|start_header_id|>system<|end_header_id|>
157
 
 
164
 
165
  | Filename | Quant type | File Size | Description |
166
  | -------- | ---------- | --------- | ----------- |
167
+ | [Llama-3-8B-Instruct-v0.9-Q2_K.gguf](https://huggingface.co/tensorblock/Llama-3-8B-Instruct-v0.9-GGUF/blob/main/Llama-3-8B-Instruct-v0.9-Q2_K.gguf) | Q2_K | 2.961 GB | smallest, significant quality loss - not recommended for most purposes |
168
+ | [Llama-3-8B-Instruct-v0.9-Q3_K_S.gguf](https://huggingface.co/tensorblock/Llama-3-8B-Instruct-v0.9-GGUF/blob/main/Llama-3-8B-Instruct-v0.9-Q3_K_S.gguf) | Q3_K_S | 3.413 GB | very small, high quality loss |
169
+ | [Llama-3-8B-Instruct-v0.9-Q3_K_M.gguf](https://huggingface.co/tensorblock/Llama-3-8B-Instruct-v0.9-GGUF/blob/main/Llama-3-8B-Instruct-v0.9-Q3_K_M.gguf) | Q3_K_M | 3.743 GB | very small, high quality loss |
170
+ | [Llama-3-8B-Instruct-v0.9-Q3_K_L.gguf](https://huggingface.co/tensorblock/Llama-3-8B-Instruct-v0.9-GGUF/blob/main/Llama-3-8B-Instruct-v0.9-Q3_K_L.gguf) | Q3_K_L | 4.025 GB | small, substantial quality loss |
171
+ | [Llama-3-8B-Instruct-v0.9-Q4_0.gguf](https://huggingface.co/tensorblock/Llama-3-8B-Instruct-v0.9-GGUF/blob/main/Llama-3-8B-Instruct-v0.9-Q4_0.gguf) | Q4_0 | 4.341 GB | legacy; small, very high quality loss - prefer using Q3_K_M |
172
+ | [Llama-3-8B-Instruct-v0.9-Q4_K_S.gguf](https://huggingface.co/tensorblock/Llama-3-8B-Instruct-v0.9-GGUF/blob/main/Llama-3-8B-Instruct-v0.9-Q4_K_S.gguf) | Q4_K_S | 4.370 GB | small, greater quality loss |
173
+ | [Llama-3-8B-Instruct-v0.9-Q4_K_M.gguf](https://huggingface.co/tensorblock/Llama-3-8B-Instruct-v0.9-GGUF/blob/main/Llama-3-8B-Instruct-v0.9-Q4_K_M.gguf) | Q4_K_M | 4.583 GB | medium, balanced quality - recommended |
174
+ | [Llama-3-8B-Instruct-v0.9-Q5_0.gguf](https://huggingface.co/tensorblock/Llama-3-8B-Instruct-v0.9-GGUF/blob/main/Llama-3-8B-Instruct-v0.9-Q5_0.gguf) | Q5_0 | 5.215 GB | legacy; medium, balanced quality - prefer using Q4_K_M |
175
+ | [Llama-3-8B-Instruct-v0.9-Q5_K_S.gguf](https://huggingface.co/tensorblock/Llama-3-8B-Instruct-v0.9-GGUF/blob/main/Llama-3-8B-Instruct-v0.9-Q5_K_S.gguf) | Q5_K_S | 5.215 GB | large, low quality loss - recommended |
176
+ | [Llama-3-8B-Instruct-v0.9-Q5_K_M.gguf](https://huggingface.co/tensorblock/Llama-3-8B-Instruct-v0.9-GGUF/blob/main/Llama-3-8B-Instruct-v0.9-Q5_K_M.gguf) | Q5_K_M | 5.339 GB | large, very low quality loss - recommended |
177
+ | [Llama-3-8B-Instruct-v0.9-Q6_K.gguf](https://huggingface.co/tensorblock/Llama-3-8B-Instruct-v0.9-GGUF/blob/main/Llama-3-8B-Instruct-v0.9-Q6_K.gguf) | Q6_K | 6.143 GB | very large, extremely low quality loss |
178
+ | [Llama-3-8B-Instruct-v0.9-Q8_0.gguf](https://huggingface.co/tensorblock/Llama-3-8B-Instruct-v0.9-GGUF/blob/main/Llama-3-8B-Instruct-v0.9-Q8_0.gguf) | Q8_0 | 7.954 GB | very large, extremely low quality loss - not recommended |
179
 
180
 
181
  ## Downloading instruction