File size: 860 Bytes
b7034a4
 
 
 
 
 
 
a64a954
 
 
 
 
 
 
 
b5ba6ed
 
 
 
a64a954
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
---
license: other
language:
- ja
- en
base_model:
- deepseek-ai/DeepSeek-V3
datasets:
- TFMC/imatrix-dataset-for-japanese-llm
---

# DeepSeek-V3-bf16-gguf
[deepseek-aiさんが公開しているDeepSeek-V3](https://huggingface.co/deepseek-ai/DeepSeek-V3)のggufフォーマット変換版です。  

imatrixのデータは[TFMC/imatrix-dataset-for-japanese-llm](https://huggingface.co/datasets/TFMC/imatrix-dataset-for-japanese-llm)を使用して作成しました。  

## License
ライセンスは`deepseek-ai/DeepSeek-V3`に準拠します。

## Usage

```
git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build -DGGML_CUDA=ON
cmake --build build --config Release
build/bin/llama-cli -m 'IQ2_XXS/DeepSeek-V3_IQ2_XXS-00001-of-00015.gguf' -n 128 -c 128 -p 'あなたはプロの料理人です。レシピを教えて' -cnv
```