Update README.md
Browse files
README.md
CHANGED
@@ -143,7 +143,7 @@ The following clients/libraries will automatically download models for you, prov
|
|
143 |
|
144 |
### In `text-generation-webui`
|
145 |
|
146 |
-
Under Download Model, you can enter the model repo: TheBloke/Llama-2-7B-GGUF and below it, a specific filename to download, such as: llama-2-7b.
|
147 |
|
148 |
Then click Download.
|
149 |
|
@@ -158,7 +158,7 @@ pip3 install huggingface-hub>=0.17.1
|
|
158 |
Then you can download any individual model file to the current directory, at high speed, with a command like this:
|
159 |
|
160 |
```shell
|
161 |
-
huggingface-cli download TheBloke/Llama-2-7B-GGUF llama-2-7b.
|
162 |
```
|
163 |
|
164 |
<details>
|
@@ -181,7 +181,7 @@ pip3 install hf_transfer
|
|
181 |
And set environment variable `HF_HUB_ENABLE_HF_TRANSFER` to `1`:
|
182 |
|
183 |
```shell
|
184 |
-
HUGGINGFACE_HUB_ENABLE_HF_TRANSFER=1 huggingface-cli download TheBloke/Llama-2-7B-GGUF llama-2-7b.
|
185 |
```
|
186 |
|
187 |
Windows CLI users: Use `set HUGGINGFACE_HUB_ENABLE_HF_TRANSFER=1` before running the download command.
|
@@ -194,7 +194,7 @@ Windows CLI users: Use `set HUGGINGFACE_HUB_ENABLE_HF_TRANSFER=1` before running
|
|
194 |
Make sure you are using `llama.cpp` from commit [d0cee0d36d5be95a0d9088b674dbb27354107221](https://github.com/ggerganov/llama.cpp/commit/d0cee0d36d5be95a0d9088b674dbb27354107221) or later.
|
195 |
|
196 |
```shell
|
197 |
-
./main -ngl 32 -m llama-2-7b.
|
198 |
```
|
199 |
|
200 |
Change `-ngl 32` to the number of layers to offload to GPU. Remove it if you don't have GPU acceleration.
|
@@ -234,7 +234,7 @@ CT_METAL=1 pip install ctransformers>=0.2.24 --no-binary ctransformers
|
|
234 |
from ctransformers import AutoModelForCausalLM
|
235 |
|
236 |
# Set gpu_layers to the number of layers to offload to GPU. Set to 0 if no GPU acceleration is available on your system.
|
237 |
-
llm = AutoModelForCausalLM.from_pretrained("TheBloke/Llama-2-7B-GGUF", model_file="llama-2-7b.
|
238 |
|
239 |
print(llm("AI is going to"))
|
240 |
```
|
|
|
143 |
|
144 |
### In `text-generation-webui`
|
145 |
|
146 |
+
Under Download Model, you can enter the model repo: TheBloke/Llama-2-7B-GGUF and below it, a specific filename to download, such as: llama-2-7b.Q4_K_M.gguf.
|
147 |
|
148 |
Then click Download.
|
149 |
|
|
|
158 |
Then you can download any individual model file to the current directory, at high speed, with a command like this:
|
159 |
|
160 |
```shell
|
161 |
+
huggingface-cli download TheBloke/Llama-2-7B-GGUF llama-2-7b.Q4_K_M.gguf --local-dir . --local-dir-use-symlinks False
|
162 |
```
|
163 |
|
164 |
<details>
|
|
|
181 |
And set environment variable `HF_HUB_ENABLE_HF_TRANSFER` to `1`:
|
182 |
|
183 |
```shell
|
184 |
+
HUGGINGFACE_HUB_ENABLE_HF_TRANSFER=1 huggingface-cli download TheBloke/Llama-2-7B-GGUF llama-2-7b.Q4_K_M.gguf --local-dir . --local-dir-use-symlinks False
|
185 |
```
|
186 |
|
187 |
Windows CLI users: Use `set HUGGINGFACE_HUB_ENABLE_HF_TRANSFER=1` before running the download command.
|
|
|
194 |
Make sure you are using `llama.cpp` from commit [d0cee0d36d5be95a0d9088b674dbb27354107221](https://github.com/ggerganov/llama.cpp/commit/d0cee0d36d5be95a0d9088b674dbb27354107221) or later.
|
195 |
|
196 |
```shell
|
197 |
+
./main -ngl 32 -m llama-2-7b.Q4_K_M.gguf --color -c 4096 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "{prompt}"
|
198 |
```
|
199 |
|
200 |
Change `-ngl 32` to the number of layers to offload to GPU. Remove it if you don't have GPU acceleration.
|
|
|
234 |
from ctransformers import AutoModelForCausalLM
|
235 |
|
236 |
# Set gpu_layers to the number of layers to offload to GPU. Set to 0 if no GPU acceleration is available on your system.
|
237 |
+
llm = AutoModelForCausalLM.from_pretrained("TheBloke/Llama-2-7B-GGUF", model_file="llama-2-7b.Q4_K_M.gguf", model_type="llama", gpu_layers=50)
|
238 |
|
239 |
print(llm("AI is going to"))
|
240 |
```
|