Skip to content

armgit / letta

Public Imported from GitHub
0

Platform for stateful agents: AI with advanced memory that can learn and self-improve over time.

feat: update reasoning toggle test (#3901)

cthomas1855eb073909
diff --git a/letta/schemas/llm_config.py b/letta/schemas/llm_config.py
index 7933d7c6d..72d4c00e6 100644
--- a/letta/schemas/llm_config.py
+++ b/letta/schemas/llm_config.py
@@ -210,9 +210,11 @@ class LLMConfig(BaseModel):
# Handle as non-reasoner until we support summary
config.put_inner_thoughts_in_kwargs = True
config.enable_reasoner = True
-
- config.put_inner_thoughts_in_kwargs = False
- config.enable_reasoner = False
+ if config.max_reasoning_tokens == 0:
+ config.max_reasoning_tokens = 1024
+ else:
+ config.put_inner_thoughts_in_kwargs = False
+ config.enable_reasoner = False
else:
config.enable_reasoner = True
diff --git a/tests/test_providers.py b/tests/test_providers.py
index cb7a54bd6..a1d55f272 100644
--- a/tests/test_providers.py
+++ b/tests/test_providers.py
@@ -320,33 +320,33 @@ async def test_provider_llm_models_consistency():
@pytest.mark.parametrize(
- "handle, expected_enable_reasoner, expected_put_inner_thoughts_in_kwargs, expected_max_reasoning_tokens, expected_reasoning_effort, expected_exception",
+ "handle, reasoning, expected_enable_reasoner, expected_put_inner_thoughts_in_kwargs, expected_max_reasoning_tokens, expected_reasoning_effort",
[
- ("openai/gpt-4o-mini", True, True, 0, None, None),
- ("openai/gpt-4o-mini", False, False, 0, None, None),
- ("openai/o3-mini", True, False, 0, "medium", None),
- ("openai/o3-mini", False, False, 0, None, ValueError),
- ("anthropic/claude-3.5-sonnet", True, True, 0, None, None),
- ("anthropic/claude-3.5-sonnet", False, False, 0, None, None),
- ("anthropic/claude-3-7-sonnet", True, False, 1024, None, None),
- ("anthropic/claude-3-7-sonnet", False, False, 0, None, None),
- ("anthropic/claude-sonnet-4", True, False, 1024, None, None),
- ("anthropic/claude-sonnet-4", False, False, 0, None, None),
- ("google_vertex/gemini-2.0-flash", True, True, 0, None, None),
- ("google_vertex/gemini-2.0-flash", False, False, 0, None, None),
- ("google_vertex/gemini-2.5-flash", True, True, 1024, None, None),
- ("google_vertex/gemini-2.5-flash", False, False, 0, None, None),
- ("google_vertex/gemini-2.5-pro", True, True, 1024, None, None),
- ("google_vertex/gemini-2.5-pro", False, False, 0, None, ValueError),
+ ("openai/gpt-4o-mini", True, True, True, 0, None),
+ ("openai/gpt-4o-mini", False, False, False, 0, None),
+ ("openai/o3-mini", True, True, False, 0, "medium"),
+ ("openai/o3-mini", False, True, False, 0, "medium"),
+ ("anthropic/claude-3.5-sonnet", True, True, True, 0, None),
+ ("anthropic/claude-3.5-sonnet", False, False, False, 0, None),
+ ("anthropic/claude-3-7-sonnet", True, True, False, 1024, None),
+ ("anthropic/claude-3-7-sonnet", False, False, False, 0, None),
+ ("anthropic/claude-sonnet-4", True, True, False, 1024, None),
+ ("anthropic/claude-sonnet-4", False, False, False, 0, None),
+ ("google_vertex/gemini-2.0-flash", True, True, True, 0, None),
+ ("google_vertex/gemini-2.0-flash", False, False, False, 0, None),
+ ("google_vertex/gemini-2.5-flash", True, True, True, 1024, None),
+ ("google_vertex/gemini-2.5-flash", False, False, False, 0, None),
+ ("google_vertex/gemini-2.5-pro", True, True, True, 1024, None),
+ ("google_vertex/gemini-2.5-pro", True, True, True, 1024, None),
],
)
def test_reasoning_toggle_by_provider(
handle: str,
+ reasoning: bool,
expected_enable_reasoner: bool,
expected_put_inner_thoughts_in_kwargs: bool,
expected_max_reasoning_tokens: int,
expected_reasoning_effort: Optional[Literal["minimal", "low", "medium", "high"]],
- expected_exception: Optional[Exception],
):
model_endpoint_type, model = handle.split("/")
config = LLMConfig(
@@ -355,13 +355,9 @@ def test_reasoning_toggle_by_provider(
handle=handle,
context_window=1024,
)
- if expected_exception:
- with pytest.raises(expected_exception):
- LLMConfig.apply_reasoning_setting_to_config(config, reasoning=expected_enable_reasoner)
- else:
- new_config = LLMConfig.apply_reasoning_setting_to_config(config, reasoning=expected_enable_reasoner)
-
- assert new_config.enable_reasoner == expected_enable_reasoner
- assert new_config.put_inner_thoughts_in_kwargs == expected_put_inner_thoughts_in_kwargs
- assert new_config.reasoning_effort == expected_reasoning_effort
- assert new_config.max_reasoning_tokens == expected_max_reasoning_tokens
+ new_config = LLMConfig.apply_reasoning_setting_to_config(config, reasoning=reasoning)
+
+ assert new_config.enable_reasoner == expected_enable_reasoner
+ assert new_config.put_inner_thoughts_in_kwargs == expected_put_inner_thoughts_in_kwargs
+ assert new_config.reasoning_effort == expected_reasoning_effort
+ assert new_config.max_reasoning_tokens == expected_max_reasoning_tokens