From 6c27754455b384a267bae21923a4f50a28d9d6f6 Mon Sep 17 00:00:00 2001 From: mateo-berri <277851410+mateo-berri@users.noreply.github.com> Date: Fri, 4 Sep 2026 09:26:51 -0700 Subject: [PATCH] fix(anthropic): bill an uncostable partial pass-through stream at zero cost instead of dropping its usage --- .../anthropic_passthrough_logging_handler.py | 22 ++++++++++++++----- ...t_anthropic_passthrough_logging_handler.py | 13 +++++++++++ 2 files changed, 29 insertions(+), 6 deletions(-) diff --git a/litellm/proxy/pass_through_endpoints/llm_provider_handlers/anthropic_passthrough_logging_handler.py b/litellm/proxy/pass_through_endpoints/llm_provider_handlers/anthropic_passthrough_logging_handler.py index dae52bab956..30b75a7b482 100644 --- a/litellm/proxy/pass_through_endpoints/llm_provider_handlers/anthropic_passthrough_logging_handler.py +++ b/litellm/proxy/pass_through_endpoints/llm_provider_handlers/anthropic_passthrough_logging_handler.py @@ -240,18 +240,28 @@ class AnthropicPassthroughLoggingHandler: usage: Final = cast(Usage | None, getattr(partial_response, "usage", None)) if partial_response is None or usage is None: return + litellm_logging_obj.record_partial_usage_for_failure( + usage=usage, + response_cost=AnthropicPassthroughLoggingHandler._cost_partial_stream_or_zero( + partial_response=partial_response, model=model, logging_obj=litellm_logging_obj + ), + ) + + @staticmethod + def _cost_partial_stream_or_zero( + partial_response: ModelResponse | TextCompletionResponse, model: str, logging_obj: LiteLLMLoggingObj + ) -> float: try: - response_cost: Final = AnthropicPassthroughLoggingHandler._compute_response_cost( + return AnthropicPassthroughLoggingHandler._compute_response_cost( litellm_model_response=partial_response, - model=AnthropicPassthroughLoggingHandler._resolve_costing_model(model, litellm_logging_obj), - logging_obj=litellm_logging_obj, + model=AnthropicPassthroughLoggingHandler._resolve_costing_model(model, logging_obj), + logging_obj=logging_obj, ) - except Exception as e: # noqa: BLE001 # an uncostable partial stream must still log as a failure + except Exception as e: # noqa: BLE001 # an uncostable partial stream still bills its tokens, at zero cost verbose_proxy_logger.warning( "Anthropic passthrough: could not cost the partial usage of a failed stream (model=%s): %s", model, e ) - return - litellm_logging_obj.record_partial_usage_for_failure(usage=usage, response_cost=response_cost) + return 0.0 @staticmethod def _compute_response_cost( diff --git a/tests/test_litellm/proxy/pass_through_endpoints/llm_provider_handlers/test_anthropic_passthrough_logging_handler.py b/tests/test_litellm/proxy/pass_through_endpoints/llm_provider_handlers/test_anthropic_passthrough_logging_handler.py index 480da1d040e..d721be62efe 100644 --- a/tests/test_litellm/proxy/pass_through_endpoints/llm_provider_handlers/test_anthropic_passthrough_logging_handler.py +++ b/tests/test_litellm/proxy/pass_through_endpoints/llm_provider_handlers/test_anthropic_passthrough_logging_handler.py @@ -2505,6 +2505,19 @@ class TestRecordPartialUsageForFailure: assert usage.prompt_tokens == 52 assert logging_obj.model_call_details["response_cost"] > 0 + def test_stashes_partial_usage_at_zero_cost_when_model_is_unpriced(self): + logging_obj = self._make_logging_obj() + + AnthropicPassthroughLoggingHandler.record_partial_usage_for_failure( + litellm_logging_obj=logging_obj, + request_body={"model": "claude-unpriced-test-model", "stream": True}, + all_chunks=self._interrupted_chunks(), + ) + + usage = logging_obj.model_call_details["combined_usage_object"] + assert usage.prompt_tokens == 52 + assert logging_obj.model_call_details["response_cost"] == 0.0 + def test_leaves_logging_obj_untouched_when_nothing_streamed(self): logging_obj = self._make_logging_obj()