Skip to content

Commit a29dc92

Browse files
author
ochafik
committed
fix server test_tool_calls.py
1 parent e1bff8f commit a29dc92

File tree

1 file changed

+29
-30
lines changed

1 file changed

+29
-30
lines changed

‎examples/server/tests/unit/test_tool_call.py‎

Lines changed: 29 additions & 30 deletions
Original file line numberDiff line numberDiff line change
@@ -274,44 +274,43 @@ def test_completion_without_tool_call_slow(template_name: str, n_predict: int, t
274274

275275

276276
@pytest.mark.slow
277-
@pytest.mark.parametrize("reasoning_format,hf_repo,template_override", [
278-
('deepseek', "bartowski/c4ai-command-r7b-12-2024-GGUF:Q4_K_M", ("CohereForAI/c4ai-command-r7b-12-2024", "tool_use")),
277+
@pytest.mark.parametrize("hf_repo,template_override", [
278+
("bartowski/Meta-Llama-3.1-8B-Instruct-GGUF:Q4_K_M", None),
279+
("bartowski/Meta-Llama-3.1-8B-Instruct-GGUF:Q4_K_M", "chatml"),
279280
280-
(None, "bartowski/Meta-Llama-3.1-8B-Instruct-GGUF:Q4_K_M", None),
281-
(None, "bartowski/Meta-Llama-3.1-8B-Instruct-GGUF:Q4_K_M", "chatml"),
281+
("bartowski/Phi-3.5-mini-instruct-GGUF:Q4_K_M", None),
282+
("bartowski/Phi-3.5-mini-instruct-GGUF:Q4_K_M", "chatml"),
282283
283-
(None, "bartowski/Phi-3.5-mini-instruct-GGUF:Q4_K_M", None),
284-
(None, "bartowski/Phi-3.5-mini-instruct-GGUF:Q4_K_M", "chatml"),
284+
("bartowski/Qwen2.5-7B-Instruct-GGUF:Q4_K_M", None),
285+
("bartowski/Qwen2.5-7B-Instruct-GGUF:Q4_K_M", "chatml"),
285286
286-
(None, "bartowski/Qwen2.5-7B-Instruct-GGUF:Q4_K_M", None),
287-
(None, "bartowski/Qwen2.5-7B-Instruct-GGUF:Q4_K_M", "chatml"),
287+
("bartowski/Hermes-2-Pro-Llama-3-8B-GGUF:Q4_K_M", ("NousResearch/Hermes-2-Pro-Llama-3-8B", "tool_use")),
288+
("bartowski/Hermes-2-Pro-Llama-3-8B-GGUF:Q4_K_M", "chatml"),
288289
289-
(None, "bartowski/Hermes-2-Pro-Llama-3-8B-GGUF:Q4_K_M", ("NousResearch/Hermes-2-Pro-Llama-3-8B", "tool_use")),
290-
(None, "bartowski/Hermes-2-Pro-Llama-3-8B-GGUF:Q4_K_M", "chatml"),
290+
("bartowski/Hermes-3-Llama-3.1-8B-GGUF:Q4_K_M", ("NousResearch/Hermes-3-Llama-3.1-8B", "tool_use")),
291+
("bartowski/Hermes-3-Llama-3.1-8B-GGUF:Q4_K_M", "chatml"),
291292
292-
(None, "bartowski/Hermes-3-Llama-3.1-8B-GGUF:Q4_K_M", ("NousResearch/Hermes-3-Llama-3.1-8B", "tool_use")),
293-
(None, "bartowski/Hermes-3-Llama-3.1-8B-GGUF:Q4_K_M", "chatml"),
293+
("bartowski/Mistral-Nemo-Instruct-2407-GGUF:Q4_K_M", None),
294+
("bartowski/Mistral-Nemo-Instruct-2407-GGUF:Q4_K_M", "chatml"),
294295
295-
(None, "bartowski/Mistral-Nemo-Instruct-2407-GGUF:Q4_K_M", None),
296-
(None, "bartowski/Mistral-Nemo-Instruct-2407-GGUF:Q4_K_M", "chatml"),
296+
("bartowski/functionary-small-v3.2-GGUF:Q8_0", ("meetkai/functionary-medium-v3.2", None)),
297+
("bartowski/functionary-small-v3.2-GGUF:Q8_0", "chatml"),
297298
298-
(None, "bartowski/functionary-small-v3.2-GGUF:Q8_0", ("meetkai/functionary-medium-v3.2", None)),
299-
(None, "bartowski/functionary-small-v3.2-GGUF:Q8_0", "chatml"),
299+
("bartowski/Llama-3.2-3B-Instruct-GGUF:Q4_K_M", ("meta-llama/Llama-3.2-3B-Instruct", None)),
300+
("bartowski/Llama-3.2-3B-Instruct-GGUF:Q4_K_M", "chatml"),
300301
301-
(None, "bartowski/Llama-3.2-3B-Instruct-GGUF:Q4_K_M", ("meta-llama/Llama-3.2-3B-Instruct", None)),
302-
(None, "bartowski/Llama-3.2-3B-Instruct-GGUF:Q4_K_M", "chatml"),
302+
("bartowski/c4ai-command-r7b-12-2024-GGUF:Q6_K_L", ("CohereForAI/c4ai-command-r7b-12-2024", "tool_use")),
303303
304-
('deepseek', "bartowski/DeepSeek-R1-Distill-Qwen-7B-GGUF:Q4_K_M", None),
304+
("bartowski/DeepSeek-R1-Distill-Qwen-7B-GGUF:Q4_K_M", None),
305305
306306
# Note: gemma-2-2b-it knows itself as "model", not "assistant", so we don't test the ill-suited chatml on it.
307-
(None, "bartowski/gemma-2-2b-it-GGUF:Q4_K_M", None),
307+
("bartowski/gemma-2-2b-it-GGUF:Q4_K_M", None),
308308
309309
# ("bartowski/Llama-3.2-1B-Instruct-GGUF:Q4_K_M", ("meta-llama/Llama-3.2-3B-Instruct", None)),
310310
])
311-
def test_weather(reasoning_format: Literal['deepseek', 'none'] | None, hf_repo: str, template_override: Tuple[str, str | None] | None):
311+
def test_weather(hf_repo: str, template_override: str | Tuple[str, str | None] | None):
312312
global server
313313
n_predict = 512
314-
server.reasoning_format = reasoning_format
315314
server.n_slots = 1
316315
server.jinja = True
317316
server.n_ctx = 8192
@@ -441,8 +440,8 @@ def test_calc_result(result_override: str | None, n_predict: int, hf_repo: str,
441440

442441
@pytest.mark.slow
443442
@pytest.mark.parametrize("n_predict,reasoning_format,expect_content,expect_reasoning_content,hf_repo,template_override", [
444-
# (1024, 'deepseek', "^The sum of 102 and 7 is 109.*", "^The user's request is straightforward.*", "bartowski/Phi-3.5-mini-instruct-GGUF:Q4_K_M", None),
445-
# (128, None, "^The sum of 102 and 7 is 109.*", None, "bartowski/Phi-3.5-mini-instruct-GGUF:Q4_K_M", None),
443+
(1024, 'deepseek', "^The sum of 102 and 7 is 109.*", None, "bartowski/Phi-3.5-mini-instruct-GGUF:Q4_K_M", None),
444+
(128, None, "^The sum of 102 and 7 is 109.*", None, "bartowski/Phi-3.5-mini-instruct-GGUF:Q4_K_M", None),
446445
447446
(1024, 'deepseek', "To find the sum of.*", "I need to calculate the sum of 102 and 7.*", "bartowski/DeepSeek-R1-Distill-Qwen-7B-GGUF:Q4_K_M", None),
448447
(1024, 'none', "<think>\nI need[\\s\\S\\r\\n]*?</think>\nTo find.*", None, "bartowski/DeepSeek-R1-Distill-Qwen-7B-GGUF:Q4_K_M", None),
@@ -491,22 +490,22 @@ def test_thoughts(n_predict: int, reasoning_format: Literal['deepseek', 'none']
491490
@pytest.mark.slow
492491
@pytest.mark.parametrize("expected_arguments_override,hf_repo,template_override", [
493492
(None, "bartowski/DeepSeek-R1-Distill-Qwen-7B-GGUF:Q4_K_M", None),
494-
(None, "bartowski/DeepSeek-R1-Distill-Qwen-7B-GGUF:Q4_K_M", "chatml"),
493+
# (None, "bartowski/DeepSeek-R1-Distill-Qwen-7B-GGUF:Q4_K_M", "chatml"),
495494
496495
(None, "bartowski/Phi-3.5-mini-instruct-GGUF:Q4_K_M", None),
497496
(None, "bartowski/Phi-3.5-mini-instruct-GGUF:Q4_K_M", "chatml"),
498497
499498
(None, "bartowski/functionary-small-v3.2-GGUF:Q8_0", ("meetkai-functionary-medium-v3.2", None)),
500499
(None, "bartowski/functionary-small-v3.2-GGUF:Q8_0", "chatml"),
501500
502-
(None, "bartowski/Meta-Llama-3.1-8B-Instruct-GGUF:Q4_K_M", None),
503-
('{"code":"print("}', "bartowski/Meta-Llama-3.1-8B-Instruct-GGUF:Q4_K_M", "chatml"),
501+
('{"code":"print("}', "bartowski/Meta-Llama-3.1-8B-Instruct-GGUF:Q4_K_M", None),
502+
(None, "bartowski/Meta-Llama-3.1-8B-Instruct-GGUF:Q4_K_M", "chatml"),
504503
505-
('{"code":"print("}', "bartowski/Llama-3.2-1B-Instruct-GGUF:Q4_K_M", ("meta-llama-Llama-3.2-3B-Instruct", None)),
504+
(None, "bartowski/Llama-3.2-1B-Instruct-GGUF:Q4_K_M", ("meta-llama-Llama-3.2-3B-Instruct", None)),
506505
(None, "bartowski/Llama-3.2-1B-Instruct-GGUF:Q4_K_M", "chatml"),
507506
508507
('{"code":"print("}', "bartowski/Llama-3.2-3B-Instruct-GGUF:Q4_K_M", ("meta-llama-Llama-3.2-3B-Instruct", None)),
509-
('{"code":"print("}', "bartowski/Llama-3.2-3B-Instruct-GGUF:Q4_K_M", "chatml"),
508+
(None, "bartowski/Llama-3.2-3B-Instruct-GGUF:Q4_K_M", "chatml"),
510509
511510
(None, "bartowski/Qwen2.5-7B-Instruct-GGUF:Q4_K_M", None),
512511
(None, "bartowski/Qwen2.5-7B-Instruct-GGUF:Q4_K_M", "chatml"),
@@ -523,7 +522,7 @@ def test_thoughts(n_predict: int, reasoning_format: Literal['deepseek', 'none']
523522
# Note: gemma-2-2b-it knows itself as "model", not "assistant", so we don't test the ill-suited chatml on it.
524523
(None, "bartowski/gemma-2-2b-it-GGUF:Q4_K_M", None),
525524
])
526-
def test_hello_world(reasoning_format: Literal['deepseek', 'none'] | None, expected_arguments_override: str | None, hf_repo: str, template_override: str | Tuple[str, str | None] | None):
525+
def test_hello_world(expected_arguments_override: str | None, hf_repo: str, template_override: str | Tuple[str, str | None] | None):
527526
global server
528527
server.n_slots = 1
529528
server.jinja = True

0 commit comments

Comments
 (0)