Mirrors/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-11-22 08:17:58 +01:00
Code Issues Packages Projects Releases Wiki Activity

Updated Templates supported by llama_chat_apply_template (markdown)

Xuan Son Nguyen 2024-03-14 22:21:55 +01:00
parent 420b3fdda3
commit e4f77876f9
1 changed files with 9 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

18
Templates-supported-by-llama_chat_apply_template.md

@ -116,6 +116,15 @@ again<end_of_turn>
response<end_of_turn>
```
```
Usage: ./server -m ... --chat-template orion
<s>Human: hello
Assistant: </s>response</s>Human: again
Assistant: </s>response</s>
```
Additionally, we also support zephyr template (I cannot find it on huggingface, but have seen in [this list](https://github.com/ggerganov/llama.cpp/blob/c8d847d57efdc0f9bbbf881d48c645e151b36fd8/examples/server/public/promptFormats.js) )
```
@ -132,15 +141,6 @@ again<|endoftext|>
response<|endoftext|>
```
```
Usage: ./server -m ... --chat-template orion
<s>Human: hello
Assistant: </s>response</s>Human: again
Assistant: </s>response</s>
```
## Custom chat templates
Currently, it's not possible to use your own chat template with llama.cpp server's `/chat/completions`