dagbs
/

deepseek-coder-7b-base-v1.5-GGUF

Inference Endpoints

Model card Files Files and versions Community

dagbs commited on Feb 3

Commit

ef3f6a8

•

1 Parent(s): 57e0467

Update README.md

Files changed (1) hide show

README.md +3 -4

README.md CHANGED Viewed

@@ -11,9 +11,8 @@ quantized_by: dagbs
 - Model organization: [DeepSeek](https://huggingface.co/deepseek-ai)
 - Original model: [deepseek-ai/deepseek-coder-7b-base-v1.5](https://huggingface.co/deepseek-ai/deepseek-coder-7b-base-v1.5)
-Notes: There are some weird things going on with the original model and these 3 GGUF files are the only usable files available. I'm not sure how useful they are or if they work.
 F16 converted using llama.cpp convert.py with the following arguments
 * --pad-vocab
-* --vocab-type bpe

 - Model organization: [DeepSeek](https://huggingface.co/deepseek-ai)
 - Original model: [deepseek-ai/deepseek-coder-7b-base-v1.5](https://huggingface.co/deepseek-ai/deepseek-coder-7b-base-v1.5)
 F16 converted using llama.cpp convert.py with the following arguments
 * --pad-vocab
+* --vocab-type hfft
+[Note]: The files _currently_ uploaded use an incorrect covab (bfe) which may results in errors and will be re-uploaded with the suite of quants soon.