vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

tell DS R1 not to overthink (weather test)

Browse source
This commit is contained in:
ochafik 2025-02-03 02:20:03 +00:00
parent ae9d5812a7
commit 19bea4ecc3
1 changed files with 5 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

7
examples/server/tests/unit/test_tool_call.py
View file

@ -251,6 +251,7 @@ def test_completion_without_tool_call_slow(template_name: str, n_predict: int, t
@pytest.mark.slow @pytest.mark.slow
@pytest.mark.parametrize("hf_repo,template_override", [ @pytest.mark.parametrize("hf_repo,template_override", [
("bartowski/DeepSeek-R1-Distill-Qwen-7B-GGUF:Q4_K_M", None),
("bartowski/c4ai-command-r7b-12-2024-GGUF:Q4_K_M", ("CohereForAI/c4ai-command-r7b-12-2024", "tool_use")), ("bartowski/c4ai-command-r7b-12-2024-GGUF:Q4_K_M", ("CohereForAI/c4ai-command-r7b-12-2024", "tool_use")),
("bartowski/Meta-Llama-3.1-8B-Instruct-GGUF:Q4_K_M", None), ("bartowski/Meta-Llama-3.1-8B-Instruct-GGUF:Q4_K_M", None),
("bartowski/gemma-2-2b-it-GGUF:Q4_K_M", None), ("bartowski/gemma-2-2b-it-GGUF:Q4_K_M", None),
@ -266,10 +267,11 @@ def test_completion_without_tool_call_slow(template_name: str, n_predict: int, t
]) ])
def test_weather(hf_repo: str, template_override: Tuple[str, str | None] | None): def test_weather(hf_repo: str, template_override: Tuple[str, str | None] | None):
global server global server
n_predict = 512
server.n_slots = 1 server.n_slots = 1
server.jinja = True server.jinja = True
server.n_ctx = 8192 server.n_ctx = 8192
server.n_predict = 512 server.n_predict = n_predict
server.model_hf_repo = hf_repo server.model_hf_repo = hf_repo
server.model_hf_file = None server.model_hf_file = None
if template_override: if template_override:
@ -278,8 +280,9 @@ def test_weather(hf_repo: str, template_override: Tuple[str, str | None] | None)
assert os.path.exists(server.chat_template_file), f"Template file {server.chat_template_file} does not exist. Run `python scripts/get_chat_template.py {template_hf_repo} {template_variant} > {server.chat_template_file}` to download the template." assert os.path.exists(server.chat_template_file), f"Template file {server.chat_template_file} does not exist. Run `python scripts/get_chat_template.py {template_hf_repo} {template_variant} > {server.chat_template_file}` to download the template."
server.start(timeout_seconds=TIMEOUT_SERVER_START) server.start(timeout_seconds=TIMEOUT_SERVER_START)
res = server.make_request("POST", "/chat/completions", data={ res = server.make_request("POST", "/chat/completions", data={
"max_tokens": 256, "max_tokens": n_predict,
"messages": [ "messages": [
{"role": "system", "content": "You are a chatbot that uses tools/functions. Dont overthink things."},
{"role": "user", "content": "What is the weather in Istanbul?"}, {"role": "user", "content": "What is the weather in Istanbul?"},
], ],
"tools": [WEATHER_TOOL], "tools": [WEATHER_TOOL],