Response Headers¶

LLMGW includes custom response headers prefixed with x-llmgw- that provide valuable metadata about each request.

Quick Reference¶

Header	Purpose	Example
`x-llmgw-request-id`	Unique request identifier for debugging	`3cb26481-d869-4923-8093-3feb92f8d9fc`
`x-llmgw-model-id`	Actual model used (may differ due to load balancing)	`azure-us-gpt35`
`x-llmgw-cost`	Request cost in USD	`0.00012`
`x-llmgw-attempts`	Number of retry attempts	`1`
`x-llmgw-remaining-limits`	Remaining usage limits	`user/daily-tokens/9500`

Accessing Headers¶

With curl¶

Use the -i flag to see all response headers:

curl -i "https://<llmgw-deployment-url>/openai/v1/chat/completions" \
  -H "Authorization: Bearer <YOUR_LLMGW_API_KEY>" \
  -H "Content-Type: application/json" \
  -d '{"model": "gpt-4.1", "messages": [{"role": "user", "content": "What is the weather like?"}]}'

With OpenAI Client (Python)¶

import json

response = client.chat.with_raw_response.completions.create(
    model="gpt-4.1",
    messages=[{"role": "user", "content": "What is the weather like?"}]
)

# Extract LLMGW headers
llmgw_headers = {
    key: value
    for key, value in response.headers.items()
    if key.startswith('x-llmgw')
}
print(json.dumps(llmgw_headers, indent=2))

With Langchain¶

from langchain_openai import AzureChatOpenAI

model = AzureChatOpenAI(
    # ... configuration
    include_response_headers=True
)

response = model.invoke("What is the weather like?")

if hasattr(response, 'response_metadata'):
    headers = response.response_metadata.get('headers', {})
    llmgw_headers = {
        key: value
        for key, value in headers.items()
        if key.startswith('x-llmgw')
    }
    print(llmgw_headers)

Example Response¶

{
  "x-llmgw-request-id": "3cb26481-d869-4923-8093-3feb92f8d9fc",
  "x-llmgw-model-id": "azure-us-gpt35",
  "x-llmgw-attempts": "1",
  "x-llmgw-cost": "0.00012",
  "x-llmgw-remaining-limits": "user/daily-requests/4999"
}

Use Cases¶

Cost Tracking: Monitor spending with x-llmgw-cost
Debugging: Use x-llmgw-request-id to trace requests in logs
Load Balancing: Check x-llmgw-model-id to see which model actually processed your request
Reliability: Monitor x-llmgw-attempts to understand retry patterns
Usage Management: Track remaining limits with x-llmgw-remaining-limits

Notes¶

Headers are returned for both successful and failed requests (when possible)
x-llmgw-remaining-limits may appear multiple times if multiple limits apply
All cost values are in USD