lianggaoquan
commited on
Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,53 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: deepseek-ai/deepseek-coder-1.3b-instruct
|
| 3 |
+
license: other
|
| 4 |
+
license_name: deepseek
|
| 5 |
+
license_link: LICENSE
|
| 6 |
+
tags:
|
| 7 |
+
- llama-cpp
|
| 8 |
+
- gguf-my-repo
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
# lianggq/deepseek-coder-1.3b-instruct-Q6_K-GGUF
|
| 12 |
+
This model was converted to GGUF format from [`deepseek-ai/deepseek-coder-1.3b-instruct`](https://huggingface.co/deepseek-ai/deepseek-coder-1.3b-instruct) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
|
| 13 |
+
Refer to the [original model card](https://huggingface.co/deepseek-ai/deepseek-coder-1.3b-instruct) for more details on the model.
|
| 14 |
+
|
| 15 |
+
## Use with llama.cpp
|
| 16 |
+
Install llama.cpp through brew (works on Mac and Linux)
|
| 17 |
+
|
| 18 |
+
```bash
|
| 19 |
+
brew install llama.cpp
|
| 20 |
+
|
| 21 |
+
```
|
| 22 |
+
Invoke the llama.cpp server or the CLI.
|
| 23 |
+
|
| 24 |
+
### CLI:
|
| 25 |
+
```bash
|
| 26 |
+
llama-cli --hf-repo lianggq/deepseek-coder-1.3b-instruct-Q6_K-GGUF --hf-file deepseek-coder-1.3b-instruct-q6_k.gguf -p "The meaning to life and the universe is"
|
| 27 |
+
```
|
| 28 |
+
|
| 29 |
+
### Server:
|
| 30 |
+
```bash
|
| 31 |
+
llama-server --hf-repo lianggq/deepseek-coder-1.3b-instruct-Q6_K-GGUF --hf-file deepseek-coder-1.3b-instruct-q6_k.gguf -c 2048
|
| 32 |
+
```
|
| 33 |
+
|
| 34 |
+
Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.
|
| 35 |
+
|
| 36 |
+
Step 1: Clone llama.cpp from GitHub.
|
| 37 |
+
```
|
| 38 |
+
git clone https://github.com/ggerganov/llama.cpp
|
| 39 |
+
```
|
| 40 |
+
|
| 41 |
+
Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
|
| 42 |
+
```
|
| 43 |
+
cd llama.cpp && LLAMA_CURL=1 make
|
| 44 |
+
```
|
| 45 |
+
|
| 46 |
+
Step 3: Run inference through the main binary.
|
| 47 |
+
```
|
| 48 |
+
./llama-cli --hf-repo lianggq/deepseek-coder-1.3b-instruct-Q6_K-GGUF --hf-file deepseek-coder-1.3b-instruct-q6_k.gguf -p "The meaning to life and the universe is"
|
| 49 |
+
```
|
| 50 |
+
or
|
| 51 |
+
```
|
| 52 |
+
./llama-server --hf-repo lianggq/deepseek-coder-1.3b-instruct-Q6_K-GGUF --hf-file deepseek-coder-1.3b-instruct-q6_k.gguf -c 2048
|
| 53 |
+
```
|