scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-10-16 04:29:31 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 139

chore: finish main merge conflict resolution for batch line items

Browse source 
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
This commit is contained in:
yucheng 2026-10-07 00:13:38 +00:00
parent 15fa86c8ec
commit 1b63787eaa
2 changed files with 53 additions and 4 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

8
litellm/batches/batch_utils.py
View file

@ -464,7 +464,9 @@ def _provider_output_file_id(output_file_id: str) -> str:
async def _fetch_batch_managed_file_content(
file_id: str,
custom_llm_provider: Literal["openai", "azure", "vertex_ai", "hosted_vllm", "anthropic", "bedrock", "mistral"] = "openai",
custom_llm_provider: Literal[
"openai", "azure", "vertex_ai", "hosted_vllm", "anthropic", "bedrock", "mistral"
] = "openai",
litellm_params: Mapping[str, object] | None = None,
) -> bytes:
"""
@ -495,7 +497,9 @@ async def _fetch_batch_managed_file_content(
async def _fetch_batch_output_file_content(
batch: Batch,
custom_llm_provider: Literal["openai", "azure", "vertex_ai", "hosted_vllm", "anthropic", "bedrock", "mistral"] = "openai",
custom_llm_provider: Literal[
"openai", "azure", "vertex_ai", "hosted_vllm", "anthropic", "bedrock", "mistral"
] = "openai",
litellm_params: Mapping[str, object] | None = None,
) -> bytes:
"""

49
tests/unit/test_router/test_router.py
View file

@ -1246,8 +1246,53 @@ def test_sync_deployment_callback_on_success_skips_batch_retrieves(
== expected_successes
)
get_parent_otel_span_from_kwargs,
is_batch_line_item_event,
@pytest.mark.asyncio
async def test_deployment_callbacks_skip_batch_line_items(monkeypatch: pytest.MonkeyPatch):
"""
Batch line-item callbacks carry call_type=acompletion plus
litellm_params.batch_parent_id, so the call-type-only batch guards do not fire.
They describe historical batch traffic already reported by the aggregate
aretrieve_batch event and must not consume live TPM/RPM quota.
"""
monkeypatch.setattr(litellm, "store_batch_line_items_in_callbacks", True, raising=False)
router = litellm.Router(
model_list=[
{
"model_name": _BATCH_GROUP,
"litellm_params": {"model": _BATCH_DEPLOYMENT_MODEL, "api_base": _BATCH_API_BASE, "api_key": "sk-fake"},
"model_info": {"id": "batch-dep"},
}
]
)
now = datetime.now()
line_item_kwargs = {
"call_type": "acompletion",
"standard_logging_object": {"total_tokens": _BATCH_TOKENS_PER_ROW},
"litellm_params": {
"batch_parent_id": _BATCH_ID,
"metadata": {"model_group": _BATCH_GROUP, "deployment": _BATCH_DEPLOYMENT_MODEL},
"model_info": {"id": "batch-dep"},
},
}
await router.deployment_callback_on_success(
kwargs=line_item_kwargs, completion_response=None, start_time=now, end_time=now
)
sync_key = router.sync_deployment_callback_on_success(
kwargs=line_item_kwargs, completion_response=None, start_time=now, end_time=now
)
await router.async_deployment_callback_on_failure(
kwargs=line_item_kwargs, completion_response=None, start_time=now, end_time=now
)
assert sync_key is None
assert (
get_deployment_successes_for_current_minute(litellm_router_instance=router, deployment_id="batch-dep") == 0
)
assert await _moved_routing_counters(router) == []
assert await _router_usage_keys(router) == []
_ROUTING_STRATEGY_CACHE_MARKERS = ("_map", "_request_count", ":tpm:", ":rpm:")