Model Usage & Cost Tracking

Across Langfuse, usage and cost are tracked for LLM generations:

• Usage: token/character counts
• Cost: USD cost of the generation

Both usage and cost can be either

• ingested via API, SDKs or integrations
• or inferred based on the model parameter of the generation. Langfuse comes with a list of predefined popular models and their tokenizers including OpenAI, Anthropic, and Google models. You can also add your own custom model definitions or request official support for new models via GitHub. Inferred cost are calculated at the time of ingestion.

Ingested usage and cost are prioritized over inferred usage and cost:

Via the Daily Metrics API, you can retrieve aggregated daily usage and cost metrics from Langfuse for downstream use in analytics, billing, and rate-limiting. The API allows you to filter by application type, user, or tags.

Ingest usage and/or cost

If available in the LLM response, ingesting usage and/or cost is the most accurate and robust way to track usage in Langfuse.

Many of the Langfuse integrations automatically capture usage and cost data from the LLM response. If this does not work as expected, please create an issue on GitHub.

@observe(as_type="generation")
def anthropic_completion(**kwargs):
  # optional, extract some fields from kwargs
  kwargs_clone = kwargs.copy()
  input = kwargs_clone.pop('messages', None)
  model = kwargs_clone.pop('model', None)
  langfuse_context.update_current_observation(
      input=input,
      model=model,
      metadata=kwargs_clone
  )
 
  response = anthopic_client.messages.create(**kwargs)
 
  langfuse_context.update_current_observation(
      usage={
          "input": response.usage.input_tokens,
          "output": response.usage.output_tokens,
          # "total": int,  # if not set, it is derived from input + output
          "unit": "TOKENS", # any of: "TOKENS", "CHARACTERS", "MILLISECONDS", "SECONDS", "IMAGES"
 
          # Optionally, also ingest usd cost. Alternatively, you can infer it via a model definition in Langfuse.
          # Here we assume the input and output cost are 1 USD each.
          "input_cost": 1,
          "output_cost": 1,
          # "total_cost": float, # if not set, it is derived from input_cost + output_cost
      }
  )
 
  # return result
  return response.content[0].text
 
@observe()
def main():
  return anthropic_completion(
      model="claude-3-opus-20240229",
      max_tokens=1024,
      messages=[
          {"role": "user", "content": "Hello, Claude"}
      ]
  )
 
main()

generation = langfuse.generation(
  # ...
  usage={
    # usage
    "input": int,
    "output": int,
    "total": int,  # if not set, it is derived from input + output
    "unit": "TOKENS",  # any of: "TOKENS", "CHARACTERS", "MILLISECONDS", "SECONDS", "IMAGES"
 
    # usd cost
    "input_cost": float,
    "output_cost": float,
    "total_cost": float, # if not set, it is derived from input_cost + output_cost
  },
  # ...
)

const generation = langfuse.generation({
  // ...
  usage: {
    // usage
    input: int,
    output: int,
    total: int, // optional, it is derived from input + output
    unit: "TOKENS", // "TOKENS" | "CHARACTERS" | "MILLISECONDS" | "SECONDS" | "IMAGES"
 
    // usd cost
    inputCost: float,
    outputCost: float,
    totalCost: float, // optional, it is derived from input + output
  },
  // ...
});

You can also update the usage and cost via generation.update() and generation.end().

Compatibility with OpenAI

For increased compatibility with OpenAI, you can also use the following attributes to ingest usage:

generation = langfuse.generation(
  # ...
  usage={
    # usage
    "prompt_tokens": int,
    "completion_tokens": int,
    "total_tokens": int,  # optional, it is derived from prompt + completion
  },
  # ...
)

const generation = langfuse.generation({
  // ...
  usage: {
    // usage
    promptTokens: integer,
    completionTokens: integer,
    totalTokens: integer, // optional, derived from prompt + completion
  },
  // ...
});

You can also ingest OpenAI-style usage via generation.update() and generation.end().

Infer usage and/or cost

If either usage or cost are not ingested, Langfuse will attempt to infer the missing values based on the model parameter of the generation at the time of ingestion. This is especially useful for some model providers or self-hosted models which do not include usage or cost in the response.

Langfuse comes with a list of predefined popular models and their tokenizers including OpenAI, Anthropic, Google. Check out the full list (you need to sign-in).

You can also add your own custom model definitions (see below) or request official support for new models via GitHub.

Usage

If a tokenizer is specified for the model, Langfuse automatically calculates token amounts for ingested generations.

The following tokenizers are currently supported:

Cost

Model definitions include prices per unit (input, output, total).

Langfuse automatically calculates cost for ingested generations at the time of ingestion if (1) usage is ingested or inferred, (2) and a matching model definition includes prices.

Custom model definitions

You can flexibly add your own model definitions to Langfuse. This is especially useful for self-hosted or fine-tuned models which are not included in the list of Langfuse maintained models.

UIAPI

Models are matched to generations based on:

User-defined models take priority over models maintained by Langfuse.

Further details

When using the openai tokenizer, you need to specify the following tokenization config. You can also copy the config from the list of predefined OpenAI models. See the OpenAI documentation for further details. tokensPerName and tokensPerMessage are required for chat models.

{
  "tokenizerModel": "gpt-3.5-turbo", // tiktoken model name
  "tokensPerName": -1, // OpenAI Chatmessage tokenization config
  "tokensPerMessage": 4 // OpenAI Chatmessage tokenization config
}

Reasoning models

Cost inference is not supported for reasoning models such as the OpenAI o1 model family. That is, if no token counts are ingested, Langfuse can not infer cost for reasoning models.

Reasoning models take multiple steps to arrive to a response. The result from each step generates reasoning tokens that are billed as output tokens. So the cost-effective output token count is the sum of all reasoning tokens and the token count for the final completion. Since Langfuse does not have visibility into the reasoning tokens, it cannot infer the correct cost for generations that have no token usage provided.

To benefit from Langfuse cost tracking, please provide the token usage when ingesting o1 model generations via the low-level SDKs. When utilizing the Langfuse OpenAI wrapper or integrations such as for Langchain, LlamaIndex or LiteLLM, token usage is collected and provided automatically for you.

For more details, see the OpenAI guide on how reasoning models work.

Troubleshooting

Usage and cost are missing for historical generations. Except for changes in prices, Langfuse does not retroactively infer usage and cost for existing generations when model definitions are changed. You can request a batch job (Langfuse Cloud) or run a script (self-hosting) to apply new model definitions to existing generations.

GitHub Discussions