langchain Architecture Docs
Home / Function/ _create_usage_metadata() — langchain Function Reference

_create_usage_metadata() — langchain Function Reference

Architecture documentation for the _create_usage_metadata() function in base.py from the langchain codebase.

Function python LangChainCore LanguageModelBase called by 2

Entity Profile

Dependency Diagram

graph TD
  81a7400d_9d67_48c6_dcb8_b344f5cba3ee["_create_usage_metadata()"]
  2b046911_ea21_8e2e_ba0d_9d03da8d7bda["base.py"]
  81a7400d_9d67_48c6_dcb8_b344f5cba3ee -->|defined in| 2b046911_ea21_8e2e_ba0d_9d03da8d7bda
  9dd73ff5_bb27_7bf2_5124_b82e93cd60f6["_convert_chunk_to_generation_chunk()"]
  9dd73ff5_bb27_7bf2_5124_b82e93cd60f6 -->|calls| 81a7400d_9d67_48c6_dcb8_b344f5cba3ee
  64924e61_d0ac_6bbf_4818_aeb51576556d["_create_chat_result()"]
  64924e61_d0ac_6bbf_4818_aeb51576556d -->|calls| 81a7400d_9d67_48c6_dcb8_b344f5cba3ee
  style 81a7400d_9d67_48c6_dcb8_b344f5cba3ee fill:#6366f1,stroke:#818cf8,color:#fff

Relationship Graph

Source Code

libs/partners/openai/langchain_openai/chat_models/base.py lines 3723–3767

def _create_usage_metadata(
    oai_token_usage: dict, service_tier: str | None = None
) -> UsageMetadata:
    input_tokens = oai_token_usage.get("prompt_tokens") or 0
    output_tokens = oai_token_usage.get("completion_tokens") or 0
    total_tokens = oai_token_usage.get("total_tokens") or input_tokens + output_tokens
    if service_tier not in {"priority", "flex"}:
        service_tier = None
    service_tier_prefix = f"{service_tier}_" if service_tier else ""
    input_token_details: dict = {
        "audio": (oai_token_usage.get("prompt_tokens_details") or {}).get(
            "audio_tokens"
        ),
        f"{service_tier_prefix}cache_read": (
            oai_token_usage.get("prompt_tokens_details") or {}
        ).get("cached_tokens"),
    }
    output_token_details: dict = {
        "audio": (oai_token_usage.get("completion_tokens_details") or {}).get(
            "audio_tokens"
        ),
        f"{service_tier_prefix}reasoning": (
            oai_token_usage.get("completion_tokens_details") or {}
        ).get("reasoning_tokens"),
    }
    if service_tier is not None:
        # Avoid counting cache and reasoning tokens towards the service tier token
        # counts, since service tier tokens are already priced differently
        input_token_details[service_tier] = input_tokens - input_token_details.get(
            f"{service_tier_prefix}cache_read", 0
        )
        output_token_details[service_tier] = output_tokens - output_token_details.get(
            f"{service_tier_prefix}reasoning", 0
        )
    return UsageMetadata(
        input_tokens=input_tokens,
        output_tokens=output_tokens,
        total_tokens=total_tokens,
        input_token_details=InputTokenDetails(
            **{k: v for k, v in input_token_details.items() if v is not None}
        ),
        output_token_details=OutputTokenDetails(
            **{k: v for k, v in output_token_details.items() if v is not None}
        ),
    )

Domain

Subdomains

Defined In

Called By

Source

Frequently Asked Questions

What does _create_usage_metadata() do?
_create_usage_metadata() is a function in the langchain codebase, defined in libs/partners/openai/langchain_openai/chat_models/base.py.
Where is _create_usage_metadata() defined?
_create_usage_metadata() is defined in libs/partners/openai/langchain_openai/chat_models/base.py at line 3723.
What calls _create_usage_metadata()?
_create_usage_metadata() is called by 2 function(s): _convert_chunk_to_generation_chunk, _create_chat_result.

Analyze Your Own Codebase

Get architecture documentation, dependency graphs, and domain analysis for your codebase in minutes.

Try Supermodel Free