StarFox7 commited on
Commit
153758f
โ€ข
1 Parent(s): 844596b

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -0
README.md CHANGED
@@ -10,6 +10,7 @@ Llama-2-ko-7V-chat-ggml ์€ [beomi/llama-2-ko-7b](https://huggingface.co/beomi/l
10
 
11
  - Llama2 tokenizer ์— [kfkas/Llama-2-ko-7b-Chat](https://huggingface.co/kfkas/Llama-2-ko-7b-Chat) ์—์„œ ์‚ฌ์šฉ๋œ ํ•œ๊ตญ์–ด Additaional Token ์„ ๋ฐ˜์˜ํ•˜์—ฌ ์ƒ์„ฑํ–ˆ์Šต๋‹ˆ๋‹ค.
12
  - **GGML** ํฌ๋งท ๋ชจ๋ธ์€ [llama.cpp](https://github.com/ggerganov/llama.cpp) ๋ฅผ ์‚ฌ์šฉํ•˜์—ฌ C/C++ ๊ธฐ๋ฐ˜์œผ๋กœ Inference ํ•ฉ๋‹ˆ๋‹ค.
 
13
  - [llama.cpp](https://github.com/ggerganov/llama.cpp) ์˜ Python Binding ํŒจํ‚ค์ง€์ธ [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) ์„ ์‚ฌ์šฉํ•˜๋ฉด python ํ™˜๊ฒฝ์—์„œ๋„ Inference ๊ฐ€๋Šฅํ•ฉ๋‹ˆ๋‹ค.
14
 
15
  ์ฐธ๊ณ ๋กœ, [Llama-2-ko-7B-ggml](https://huggingface.co/StarFox7/Llama-2-ko-7B-ggml) ์—์„œ [Llama-2-ko-7b-chat](https://huggingface.co/kfkas/Llama-2-ko-7b-Chat) ์˜ ๋ฒ ์ด์Šค๋ชจ๋ธ์ธ [beomi/llama-2-ko-7b](https://huggingface.co/beomi/llama-2-ko-7b) ์˜ **GGML** ํฌ๋งท ๋ชจ๋ธ์„ ์ฐพ์„ ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
 
10
 
11
  - Llama2 tokenizer ์— [kfkas/Llama-2-ko-7b-Chat](https://huggingface.co/kfkas/Llama-2-ko-7b-Chat) ์—์„œ ์‚ฌ์šฉ๋œ ํ•œ๊ตญ์–ด Additaional Token ์„ ๋ฐ˜์˜ํ•˜์—ฌ ์ƒ์„ฑํ–ˆ์Šต๋‹ˆ๋‹ค.
12
  - **GGML** ํฌ๋งท ๋ชจ๋ธ์€ [llama.cpp](https://github.com/ggerganov/llama.cpp) ๋ฅผ ์‚ฌ์šฉํ•˜์—ฌ C/C++ ๊ธฐ๋ฐ˜์œผ๋กœ Inference ํ•ฉ๋‹ˆ๋‹ค.
13
+ - **GGML** ํฌ๋งท ๋ชจ๋ธ์€ ๋น„๊ต์  ๋‚ฎ์€ ์‚ฌ์–‘์˜ ์ปดํ“จํŒ… ์ž์›์—์„œ๋„ Inference ๊ฐ€๋Šฅํ•ฉ๋‹ˆ๋‹ค. ( ์˜ˆ: 4๋น„ํŠธ ์–‘์žํ™” ๋ชจ๋ธ (q4) ์€ CPU,7-8GB RAM ํ™˜๊ฒฝ์—์„œ Inference ๊ฐ€๋Šฅ )
14
  - [llama.cpp](https://github.com/ggerganov/llama.cpp) ์˜ Python Binding ํŒจํ‚ค์ง€์ธ [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) ์„ ์‚ฌ์šฉํ•˜๋ฉด python ํ™˜๊ฒฝ์—์„œ๋„ Inference ๊ฐ€๋Šฅํ•ฉ๋‹ˆ๋‹ค.
15
 
16
  ์ฐธ๊ณ ๋กœ, [Llama-2-ko-7B-ggml](https://huggingface.co/StarFox7/Llama-2-ko-7B-ggml) ์—์„œ [Llama-2-ko-7b-chat](https://huggingface.co/kfkas/Llama-2-ko-7b-Chat) ์˜ ๋ฒ ์ด์Šค๋ชจ๋ธ์ธ [beomi/llama-2-ko-7b](https://huggingface.co/beomi/llama-2-ko-7b) ์˜ **GGML** ํฌ๋งท ๋ชจ๋ธ์„ ์ฐพ์„ ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.