Update Inference Result

Update an existing inference result with agent output and metadata

Overview

The update() method allows you to modify an existing inference result after it has been created. This is useful when you need to add or update output, latency, token usage, or cost information.

Parameters

inference_result_id · string · required

The ID of the inference result to update.

Output Fields

output · string

The generated output or response from the AI model.

input · string

The input text or prompt for the inference result.

retrieval_context · string

The context retrieved by a RAG system, if applicable.

When you set retrieval_context, Galtea also records it as a RETRIEVER trace on this inference result. Sending null (or an empty string) clears the value and removes that trace.

Performance Fields

latency · float

The time in milliseconds from request to response.

Usage Fields

input_tokens · int

Number of input tokens sent to the model.

output_tokens · int

Number of output tokens generated by the model.

cache_read_input_tokens · int

Number of input tokens read from the cache.

tokens · int

Total tokens used in the model call.

Cost Fields

cost · float

The total cost associated with the model call.

cost_per_input_token · float

Cost per input token sent to the model.

cost_per_output_token · float

Cost per output token generated by the model.

cost_per_cache_read_input_token · float

Cost per input token read from the cache.

Returns

Returns the updated InferenceResult object.

Example

inference_result = galtea.inference_results.update(
    inference_result_id=inference_result_id,
    output="Paris is the capital.",
    cost=0.0001,
)

Use Cases

Deferred Output Update

Create an inference result first, then update it after processing completes:

# Create inference result with just input
user_input = "What is the weather today?"
inference_result = galtea.inference_results.create(session_id=session.id, input=user_input)
if inference_result is None:
    raise ValueError("inference_result is None")

# Process with your model
start_time = time.time()
response = my_model_generate(user_input)
latency_ms = (time.time() - start_time) * 1000

# Update with output and metrics
galtea.inference_results.update(inference_result_id=inference_result.id, output=response, latency=latency_ms)

Adding Cost Information

Update an inference result with cost data after receiving billing info:

galtea.inference_results.update(
    inference_result_id=inference_result.id,
    cost=0.0025,
    cost_per_input_token=0.00001,
    cost_per_output_token=0.00003,
)

Notes

All fields except inference_result_id default to PydanticUndefined (from pydantic_core). Omit a field (or pass PydanticUndefined) to leave it unchanged. Pass None to explicitly clear an optional field. Pass a value to update it.

• The creditsUsed field cannot be modified through this method

Related Methods

• Create Inference Result - Create a new inference result
• Generate Inference Result - Create with automatic trace collection
• Get Inference Result - Retrieve an inference result