root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2025-01-13 12:10:18 +00:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

Update README.md

Browse Source
This commit is contained in:
ochafik 2024-09-27 07:18:07 +01:00
parent 86e4f99092
commit 2f25ee30ef
1 changed files with 5 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
examples/agent/README.md
View File

@ -14,17 +14,21 @@
<details>
<summary>Instructions for NousResearch/Hermes-2-Pro-Llama-3-8B (needs template override)</summary>
The HF model had two variants for its chat template (`default` and `tool_use`), but the GGUF only retained the `default` one.
```bash
./llama-server \
--jinja -fa \
-mu https://huggingface.co/NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF/resolve/main/Hermes-2-Pro-Llama-3-8B-Q8_0.gguf \
--chat-template-file tests/chat/templates/NousResearch-Hermes-2-Pro-Llama-3-8B-tool_use.jinja
```
`
</details>
<details>
<summary>Instructions for meekai/functionary-small-v3.2 (needs template override)</summary>
The template in the GGUF doesn't seem to support tool calls, but its bigger brother's template can be used:
The template in the GGUF doesn't support tool calls, but its bigger brother's template can be used:
```bash
./llama-server \