Importing GGUF to Ollama made easy
- Import GGUF files to Ollama
- Auto-generate Modelfile
- Automatic STOP detection
[INST] ... [/INST]
[SYSTEM_PROMPT] ... [/SYSTEM_PROMPT]
[AVAILABLE_TOOLS] ... [/AVAILABLE_TOOLS]
[TOOL_RESULTS] ... [/TOOL_RESULTS]
[TOOL_CALL] ... [/TOOL_CALL]
<<SYS>> ... <</SYS>>
<s> ... </s>
<start_of_turn> ... <end_of_turn>
<|start_header_id|> ... <|end_header_id|>
<｜begin▁of▁sentence｜> ... <｜begin▁of▁sentence｜>
<｜User｜>
<｜Assistant｜>
<|eot_id|>
<|im_sep|>
<|reserved_special_token
gguf_to_ollama <gguf-file> [<name>] [-context=65536]<gguf-file>
GGUF file to import to Ollama
<name>
Name of the model (optional)
-context
Context size fixed (optional, default by ollama: 2048)
If modelfile already exists, it will not be overwritten.
You can edit the modelfile manually. Documentation