mirror of
https://github.com/BerriAI/litellm.git
synced 2026-09-28 01:32:17 +00:00
5e6dc89ba1
* ci: run the unit_selection.sh shard files on every event instead of only fork pull requests Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * ci: rename fork-flag to unit-flag now that it applies on every event * test: move tests/test_litellm root and small trees into tests/unit Pure renames, no content changes. Follow-up commits in this PR fix references, merge the three files that already existed in tests/unit, keep live-provider tests in tests/test_litellm and wire CI. * test: carry tests/test_litellm conftest isolation into tests/unit Callback lists, routing fallbacks, cached HTTP clients, logger state, AWS, proxy-URL and keychain env, and session-end client cleanup now reset for unit tests too. The environment isolation owns its MonkeyPatch so a test's own monkeypatch is undone before the model-cost teardown runs. * test: merge, split and prune the moved root and small-tree tests Merge batches/test_batch_utils.py and the chat_completions and messages dispatch tests into the files that already existed in tests/unit. Keep the live Gemini interactions tests, the async image-fetch format test and the OpenAI embedding scorer test in tests/test_litellm since they need real network or keys. Put test_router.py under tests/unit/test_router so the existing package no longer shadows it. Delete eight tests the audit found superseded by stronger ones kept in this move. * ci: run the moved root and small-tree tests under their legacy flags Add the misc and responses-caching-types flags to unit_selection.sh and CircleCI, extend enterprise-routing and mcp-integration, and point the legacy GHA shards, Makefile, redis-compat workflow, merge smoke manifest and change classifier at the new paths. * test: make the new tests/unit directories packages tests/unit/test_package_layout.py requires every directory to carry an __init__.py, and without one the moved and retained test_litellm_responses_bridge.py modules collide on import. * test: scope the unit socket block to tests/unit in shared sessions The GHA shards collect the legacy test-path and the unit selection in one pytest session. The unit conftest's loopback-only block leaked into legacy modules that reach the network at import. The legacy conftest now lifts the restriction at collect and setup time, and the unit conftest re-applies it when collecting its own modules. * test: move tests/test_litellm/llms into tests/unit/llms Rename-only. Moves the provider tests and the fine-tuning fixtures they load, mirroring the old paths. Follow-up commits merge, split and wire them. * test: merge, split and prune the moved llms tests Merges the Databricks chat transformation tests into the existing unit file, keeps the tests that need real keys or the network in tests/test_litellm, deletes the audited tests a stronger unit test already covers, and points imports at tests.unit.llms. * ci: run the moved llms tests under their legacy flags The Vertex AI and All Other Providers shards keep their legacy test-path for the retained files and add the llm-vertex-ai and llm-other-providers unit selections. CircleCI gets matching unit jobs. * test: make the tests/unit/llms directories packages Adds __init__.py to the moved dirs and drops the legacy ones whose directories no longer hold tests. * test: drop script runners and path hacks the llms split left dangling The __main__ runners in the split openai_like files and the Databricks e2e runner called tests that now live in the other half of the split or were deleted. The retained legacy halves also no longer need sys.path edits. * test: give the shard-script tests their own GITHUB_OUTPUT They only passed where the runner set it. The CircleCI unit job's env allowlist drops it, so the script's redirect failed there. * test: point the router and module-deletion checks at tests/unit router_code_coverage and code_qa_check_tests only searched tests/test_litellm, so the moved router tests no longer counted. The two silent-experiment tests the audit deleted were the only direct callers of those methods; they are replaced with tests that assert the forwarded shadow request and the recursion guard. * test: keep the Databricks manual e2e runner and fix the SageMaker Nova run path The Databricks e2e file is a manual script whose main() calls the tests that were pruned, so pruning them broke the documented run. It is back to its main version. The SageMaker Nova docstring now points at the file's real location in tests/local_testing. * test: keep the job's UNIT_FLAG out of the shard-script tests --------- Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
275 lines
8.7 KiB
Python
275 lines
8.7 KiB
Python
"""
|
|
Test cache_control and reasoning parameter support for MiniMax, GLM/ZAI, and OpenRouter.
|
|
|
|
This test file verifies the fixes for Issue #19923:
|
|
- cache_control is preserved (not stripped) for MiniMax, GLM, and OpenRouter variants
|
|
- thinking parameter is supported for reasoning-capable models
|
|
- Model metadata correctly reflects capabilities
|
|
"""
|
|
|
|
|
|
import pytest
|
|
|
|
|
|
from litellm.llms.minimax.chat.transformation import MinimaxChatConfig
|
|
from litellm.llms.openrouter.chat.transformation import OpenrouterConfig
|
|
from litellm.llms.zai.chat.transformation import ZAIChatConfig
|
|
|
|
|
|
def test_minimax_preserves_cache_control_in_messages():
|
|
"""MiniMax should NOT strip cache_control from messages."""
|
|
config = MinimaxChatConfig()
|
|
|
|
messages = [
|
|
{
|
|
"role": "system",
|
|
"content": "You are a helpful assistant.",
|
|
"cache_control": {"type": "ephemeral"},
|
|
},
|
|
{
|
|
"role": "user",
|
|
"content": "Hello, world!",
|
|
},
|
|
]
|
|
|
|
transformed_messages, _ = config.remove_cache_control_flag_from_messages_and_tools(
|
|
model="minimax/MiniMax-M2.1", messages=messages
|
|
)
|
|
|
|
# cache_control should be preserved
|
|
assert transformed_messages[0].get("cache_control") == {"type": "ephemeral"}
|
|
|
|
|
|
def test_minimax_preserves_cache_control_in_tools():
|
|
"""MiniMax should NOT strip cache_control from tools."""
|
|
config = MinimaxChatConfig()
|
|
|
|
tools = [
|
|
{
|
|
"type": "function",
|
|
"function": {
|
|
"name": "get_weather",
|
|
"description": "Get weather information",
|
|
"parameters": {"type": "object", "properties": {}},
|
|
},
|
|
"cache_control": {"type": "ephemeral"},
|
|
}
|
|
]
|
|
|
|
_, transformed_tools = config.remove_cache_control_flag_from_messages_and_tools(
|
|
model="minimax/MiniMax-M2.1", messages=[], tools=tools
|
|
)
|
|
|
|
# cache_control should be preserved
|
|
assert transformed_tools[0].get("cache_control") == {"type": "ephemeral"}
|
|
|
|
|
|
def test_minimax_supports_thinking_param():
|
|
"""MiniMax reasoning models should support thinking parameter."""
|
|
config = MinimaxChatConfig()
|
|
|
|
supported_params = config.get_supported_openai_params(model="minimax/MiniMax-M2.1")
|
|
|
|
# thinking should be in supported params for reasoning models
|
|
assert "thinking" in supported_params
|
|
# reasoning_split should also be supported
|
|
assert "reasoning_split" in supported_params
|
|
|
|
|
|
def test_zai_preserves_cache_control_in_messages():
|
|
"""ZAI should NOT strip cache_control from messages."""
|
|
config = ZAIChatConfig()
|
|
|
|
messages = [
|
|
{
|
|
"role": "system",
|
|
"content": "You are a helpful assistant.",
|
|
"cache_control": {"type": "ephemeral"},
|
|
},
|
|
{
|
|
"role": "user",
|
|
"content": "Hello, world!",
|
|
},
|
|
]
|
|
|
|
transformed_messages, _ = config.remove_cache_control_flag_from_messages_and_tools(
|
|
model="zai/glm-4.7", messages=messages
|
|
)
|
|
|
|
# cache_control should be preserved
|
|
assert transformed_messages[0].get("cache_control") == {"type": "ephemeral"}
|
|
|
|
|
|
def test_zai_preserves_cache_control_in_tools():
|
|
"""ZAI should NOT strip cache_control from tools."""
|
|
config = ZAIChatConfig()
|
|
|
|
tools = [
|
|
{
|
|
"type": "function",
|
|
"function": {
|
|
"name": "get_weather",
|
|
"description": "Get weather information",
|
|
"parameters": {"type": "object", "properties": {}},
|
|
},
|
|
"cache_control": {"type": "ephemeral"},
|
|
}
|
|
]
|
|
|
|
_, transformed_tools = config.remove_cache_control_flag_from_messages_and_tools(
|
|
model="zai/glm-4.7", messages=[], tools=tools
|
|
)
|
|
|
|
# cache_control should be preserved
|
|
assert transformed_tools[0].get("cache_control") == {"type": "ephemeral"}
|
|
|
|
|
|
def test_zai_supports_thinking_param_for_reasoning_models():
|
|
"""ZAI reasoning models (glm-4.7, glm-4.6) should support thinking parameter."""
|
|
config = ZAIChatConfig()
|
|
|
|
# glm-4.7 supports reasoning
|
|
supported_params_47 = config.get_supported_openai_params(model="zai/glm-4.7")
|
|
assert "thinking" in supported_params_47
|
|
|
|
# glm-4.6 supports reasoning
|
|
supported_params_46 = config.get_supported_openai_params(model="zai/glm-4.6")
|
|
assert "thinking" in supported_params_46
|
|
|
|
|
|
def test_openrouter_minimax_supports_cache_control():
|
|
"""OpenRouter should preserve cache_control for MiniMax models."""
|
|
config = OpenrouterConfig()
|
|
|
|
messages = [
|
|
{
|
|
"role": "user",
|
|
"content": "Hello, world!",
|
|
"cache_control": {"type": "ephemeral"},
|
|
}
|
|
]
|
|
|
|
# Test that cache_control is not removed
|
|
transformed_messages, _ = config.remove_cache_control_flag_from_messages_and_tools(
|
|
model="openrouter/minimax/minimax-m2", messages=messages
|
|
)
|
|
|
|
# The method should preserve cache_control for minimax models
|
|
assert transformed_messages[0].get("cache_control") == {"type": "ephemeral"}
|
|
|
|
|
|
def test_openrouter_glm_supports_cache_control():
|
|
"""OpenRouter should preserve cache_control for GLM models."""
|
|
config = OpenrouterConfig()
|
|
|
|
messages = [
|
|
{
|
|
"role": "user",
|
|
"content": "Hello, world!",
|
|
"cache_control": {"type": "ephemeral"},
|
|
}
|
|
]
|
|
|
|
# Test that cache_control is not removed for GLM models
|
|
transformed_messages, _ = config.remove_cache_control_flag_from_messages_and_tools(
|
|
model="openrouter/z-ai/glm-4.6", messages=messages
|
|
)
|
|
|
|
# The method should preserve cache_control for GLM models
|
|
assert transformed_messages[0].get("cache_control") == {"type": "ephemeral"}
|
|
|
|
|
|
def test_openrouter_deepseek_strips_cache_control():
|
|
"""OpenRouter should still strip cache_control for non-supported models."""
|
|
config = OpenrouterConfig()
|
|
|
|
messages = [
|
|
{
|
|
"role": "user",
|
|
"content": "Hello, world!",
|
|
"cache_control": {"type": "ephemeral"},
|
|
}
|
|
]
|
|
|
|
# DeepSeek doesn't support cache_control, so it should be stripped
|
|
transformed_messages, _ = config.remove_cache_control_flag_from_messages_and_tools(
|
|
model="openrouter/deepseek/deepseek-chat", messages=messages
|
|
)
|
|
|
|
# cache_control should be removed for non-supported models
|
|
assert transformed_messages[0].get("cache_control") is None
|
|
|
|
|
|
def test_openrouter_minimax_transform_moves_cache_control_to_content():
|
|
"""OpenRouter should move cache_control to content blocks for MiniMax."""
|
|
config = OpenrouterConfig()
|
|
|
|
messages = [
|
|
{
|
|
"role": "user",
|
|
"content": "Analyze this data",
|
|
"cache_control": {"type": "ephemeral"},
|
|
}
|
|
]
|
|
|
|
transformed_request = config.transform_request(
|
|
model="openrouter/minimax/minimax-m2",
|
|
messages=messages,
|
|
optional_params={},
|
|
litellm_params={},
|
|
headers={},
|
|
)
|
|
|
|
# cache_control should be moved to content blocks
|
|
assert "messages" in transformed_request
|
|
user_message = transformed_request["messages"][0]
|
|
assert isinstance(user_message["content"], list)
|
|
assert user_message["content"][0]["cache_control"] == {"type": "ephemeral"}
|
|
# Message-level cache_control should be removed
|
|
assert "cache_control" not in user_message
|
|
|
|
|
|
def test_openrouter_glm_transform_moves_cache_control_to_content():
|
|
"""OpenRouter should move cache_control to content blocks for GLM."""
|
|
config = OpenrouterConfig()
|
|
|
|
messages = [
|
|
{
|
|
"role": "user",
|
|
"content": "Analyze this data",
|
|
"cache_control": {"type": "ephemeral"},
|
|
}
|
|
]
|
|
|
|
transformed_request = config.transform_request(
|
|
model="openrouter/z-ai/glm-4.6",
|
|
messages=messages,
|
|
optional_params={},
|
|
litellm_params={},
|
|
headers={},
|
|
)
|
|
|
|
# cache_control should be moved to content blocks
|
|
assert "messages" in transformed_request
|
|
user_message = transformed_request["messages"][0]
|
|
assert isinstance(user_message["content"], list)
|
|
assert user_message["content"][0]["cache_control"] == {"type": "ephemeral"}
|
|
|
|
|
|
def test_openrouter_supports_thinking_param_for_reasoning_models():
|
|
"""OpenRouter should support thinking parameter for reasoning-capable models."""
|
|
config = OpenrouterConfig()
|
|
|
|
# Test MiniMax (supports reasoning)
|
|
supported_params_minimax = config.get_supported_openai_params(
|
|
model="openrouter/minimax/minimax-m2"
|
|
)
|
|
assert "thinking" in supported_params_minimax
|
|
assert "reasoning_effort" in supported_params_minimax
|
|
|
|
# Test GLM (supports reasoning)
|
|
supported_params_glm = config.get_supported_openai_params(
|
|
model="openrouter/z-ai/glm-4.6"
|
|
)
|
|
assert "thinking" in supported_params_glm
|
|
assert "reasoning_effort" in supported_params_glm
|