llama-stack-mirror/llama_stack/models/llama/llama3
ehhuang 1e5bf6c19d
feat: update default tool use prompt (#1803)
# What does this PR do?
User reports in
https://github.com/meta-llama/llama-stack/issues/1769#issuecomment-2755564632
that Agent uses tool even on a prompt 'Hello'.

Updated the default prompt. Also move the instruction part out of
`function_description` so that user can override it if desired.

## Test Plan
<img width="1344" alt="image"
src="https://github.com/user-attachments/assets/c606d65d-071f-4211-a719-b4742676acda"
/>

Also performance on 100 hotpotqa questions are similar to the current
prompt.
2025-04-12 11:54:22 -07:00
..
multimodal refactor: move all llama code to models/llama out of meta reference (#1887) 2025-04-07 15:03:58 -07:00
prompt_templates feat: update default tool use prompt (#1803) 2025-04-12 11:54:22 -07:00
quantization refactor: move all llama code to models/llama out of meta reference (#1887) 2025-04-07 15:03:58 -07:00
__init__.py chore: remove dependency on llama_models completely (#1344) 2025-03-01 12:48:08 -08:00
args.py refactor: move all llama code to models/llama out of meta reference (#1887) 2025-04-07 15:03:58 -07:00
chat_format.py feat: add batch inference API to llama stack inference (#1945) 2025-04-12 11:41:12 -07:00
dog.jpg chore: move all Llama Stack types from llama-models to llama-stack (#1098) 2025-02-14 09:10:59 -08:00
generation.py feat: add batch inference API to llama stack inference (#1945) 2025-04-12 11:41:12 -07:00
interface.py refactor: move all llama code to models/llama out of meta reference (#1887) 2025-04-07 15:03:58 -07:00
model.py refactor: move all llama code to models/llama out of meta reference (#1887) 2025-04-07 15:03:58 -07:00
pasta.jpeg chore: move all Llama Stack types from llama-models to llama-stack (#1098) 2025-02-14 09:10:59 -08:00
template_data.py refactor: move all llama code to models/llama out of meta reference (#1887) 2025-04-07 15:03:58 -07:00
tokenizer.model chore: remove dependency on llama_models completely (#1344) 2025-03-01 12:48:08 -08:00
tokenizer.py refactor: move all llama code to models/llama out of meta reference (#1887) 2025-04-07 15:03:58 -07:00
tool_utils.py refactor: move all llama code to models/llama out of meta reference (#1887) 2025-04-07 15:03:58 -07:00