Skip to content

The fastest, litest AI Gateway. Rust core with Python SDK. Call 100+ LLM APIs in OpenAI (or native) format with cost tracking, guardrails, load balancing, and logging [Bedrock, Azure, OpenAI, Anthropic, OpenAI, VertexAI, vLLM, Nvidia NIM]

fix(proxy): forward dispatched_call_type through _failure_usage_to_lift (#47815)

Mateo Wanga31c5230de30
diff --git a/litellm/proxy/utils.py b/litellm/proxy/utils.py
index 52b3866d04d..28be11509ae 100644
--- a/litellm/proxy/utils.py
+++ b/litellm/proxy/utils.py
@@ -1294,6 +1294,7 @@ def _failure_usage_to_lift(
provider_billed: bool = False,
deployment_id: str | None = None,
base_model: str | None = None,
+ dispatched_call_type: str | None = None,
) -> tuple[object, object] | None:
current_attempt: Final = _current_failure_usage_to_lift(
model_call_details,
@@ -1302,6 +1303,7 @@ def _failure_usage_to_lift(
provider_billed,
deployment_id,
base_model,
+ dispatched_call_type=dispatched_call_type,
)
failed_attempts: Final = _FAILED_ATTEMPTS_USAGE.validate_python(model_call_details.get("failed_attempts_usage", ()))
if not failed_attempts: