(fix) propagate x-litellm-model-id in responses

[raghav-stripe]

Title

Propagate model ID header (x-litellm-model-id) on error responses

Relevant issues

Pre-Submission checklist

Please complete all items before asking a LiteLLM maintainer to review your PR

• I have Added testing in the tests/litellm/ directory, Adding at least 1 test is a hard requirement - see details
• I have added a screenshot of my new test passing locally
• My PR passes all unit tests on make test-unit
• My PR's scope is as isolated as possible, it only solves 1 specific problem

Type

Bug Fix

Changes

This PR ensures the x-litellm-model-id header is consistently returned on both successful and failed requests across all endpoints (/v1/chat/completions, /v1/responses, /v1/messages) for sync and streaming calls.

Previously, the model ID header was only returned on successful requests, and only for chat completions + responses API. It was missing on messages entirely, and responses API streaming as well, as well as error cases (non-200) for all the endpoints. This causes issues for clients who need to track which specific deployment handled their request, especially when debugging errors / tracking for metrics!

The fix extracts the model_id from the router's metadata (litellm_metadata) which is populated during routing and persists throughout the request lifecycle, including when errors occur after a deployment has been selected. This ensures consistent behavior across all response types (success, streaming, and error).

Test Results

Passing unit test screenshot:

Start the server:

poetry run litellm --config config.yaml --port 4000

/v1/chat/completions

Success - Non-streaming:

curl -v http://localhost:4000/v1/chat/completions \
  -H 'Authorization: Bearer sk-master' \
  -H 'Content-Type: application/json' \
  -d '{"model": "gpt-4.1", "messages": [{"role": "user", "content": "Hi"}]}' \
  2>&1 | grep -i "x-litellm-model-id"

< x-litellm-model-id: 91f84a8c74066cd29acc087803ba323645417b95f9e55096bf8e3827149e8492

Success - Streaming:

curl -v http://localhost:4000/v1/chat/completions \
  -H 'Authorization: Bearer sk-master' \
  -H 'Content-Type: application/json' \
  -d '{"model": "gpt-4.1", "messages": [{"role": "user", "content": "Hi"}], "stream": true}' \
  2>&1 | grep -i "x-litellm-model-id"

< x-litellm-model-id: 91f84a8c74066cd29acc087803ba323645417b95f9e55096bf8e3827149e8492

Error - Bad Request:

curl -v http://localhost:4000/v1/chat/completions \
  -H 'Authorization: Bearer sk-master' \
  -H 'Content-Type: application/json' \
  -d '{"model": "gpt-4.1", "messages": [{"role": "user", "content": "Hi"}], "reasoning_effort": "low"}' \
  2>&1 | grep -i "x-litellm-model-id"

< x-litellm-model-id: 91f84a8c74066cd29acc087803ba323645417b95f9e55096bf8e3827149e8492

Error - Streaming Bad Request:

curl -v http://localhost:4000/v1/chat/completions \
  -H 'Authorization: Bearer sk-master' \
  -H 'Content-Type: application/json' \
  -d '{"model": "gpt-4.1", "messages": [{"role": "user", "content": "Hi"}], "reasoning_effort": "low", "stream": true}' \
  2>&1 | grep -i "x-litellm-model-id"

< x-litellm-model-id: 91f84a8c74066cd29acc087803ba323645417b95f9e55096bf8e3827149e8492

Success - Non-streaming - Cross-provider to Bedrock:

curl -v http://localhost:4000/v1/chat/completions \
  -H 'Authorization: Bearer sk-master' \
  -H 'Content-Type: application/json' \
  -d '{"model": "claude-sonnet-4", "messages": [{"role": "user", "content": "Hi"}]}' \
  2>&1 | grep -i "x-litellm-model-id"

< x-litellm-model-id: 0d410bc56f642789e11dec790cf9c3ba46c680f1b6e25c5a9cf171bb1f2f3791

Success - Streaming - Cross-provider to Bedrock:

curl -v http://localhost:4000/v1/chat/completions \
  -H 'Authorization: Bearer sk-master' \
  -H 'Content-Type: application/json' \
  -d '{"model": "claude-sonnet-4", "messages": [{"role": "user", "content": "Hi"}], "stream": true}' \
  2>&1 | grep -i "x-litellm-model-id"

< x-litellm-model-id: 0d410bc56f642789e11dec790cf9c3ba46c680f1b6e25c5a9cf171bb1f2f3791

Error - Bad Request - Cross-provider to Bedrock:

curl -v http://localhost:4000/v1/chat/completions \
  -H 'Authorization: Bearer sk-master' \
  -H 'Content-Type: application/json' \
  -d '{"model": "claude-sonnet-4", "messages": [{"role": "user", "content": "Hi"}], "max_tokens": 0}' \
  2>&1 | grep -i "x-litellm-model-id"

< x-litellm-model-id: 0d410bc56f642789e11dec790cf9c3ba46c680f1b6e25c5a9cf171bb1f2f3791

Error - Streaming Bad Request - Cross-provider to Bedrock:

curl -v http://localhost:4000/v1/chat/completions \
  -H 'Authorization: Bearer sk-master' \
  -H 'Content-Type: application/json' \
  -d '{"model": "claude-sonnet-4", "messages": [{"role": "user", "content": "Hi"}], "max_tokens": 0, "stream": true}' \
  2>&1 | grep -i "x-litellm-model-id"

< x-litellm-model-id: 0d410bc56f642789e11dec790cf9c3ba46c680f1b6e25c5a9cf171bb1f2f3791

---

/v1/messages

Success - Non-streaming:

curl -v http://localhost:4000/v1/messages \
  -H 'Authorization: Bearer sk-master' \
  -H 'Content-Type: application/json' \
  -d '{"model": "claude-sonnet-4", "max_tokens": 10, "messages": [{"role": "user", "content": "Hi"}]}' \
  2>&1 | grep -i "x-litellm-model-id"

< x-litellm-model-id: 0d410bc56f642789e11dec790cf9c3ba46c680f1b6e25c5a9cf171bb1f2f3791

Success - Streaming:

curl -v http://localhost:4000/v1/messages \
  -H 'Authorization: Bearer sk-master' \
  -H 'Content-Type: application/json' \
  -d '{"model": "claude-sonnet-4", "max_tokens": 10, "messages": [{"role": "user", "content": "Hi"}], "stream": true}' \
  2>&1 | grep -i "x-litellm-model-id"

< x-litellm-model-id: 0d410bc56f642789e11dec790cf9c3ba46c680f1b6e25c5a9cf171bb1f2f3791

Error - Bad Request:

curl -v http://localhost:4000/v1/messages \
  -H 'Authorization: Bearer sk-master' \
  -H 'Content-Type: application/json' \
  -d '{"model": "claude-sonnet-4", "max_tokens": 0, "messages": [{"role": "user", "content": "Hi"}]}' \
  2>&1 | grep -i "x-litellm-model-id"

< x-litellm-model-id: 0d410bc56f642789e11dec790cf9c3ba46c680f1b6e25c5a9cf171bb1f2f3791

Error - Streaming Bad Request:

curl -v http://localhost:4000/v1/messages \
  -H 'Authorization: Bearer sk-master' \
  -H 'Content-Type: application/json' \
  -d '{"model": "claude-sonnet-4", "max_tokens": 0, "messages": [{"role": "user", "content": "Hi"}], "stream": true}' \
  2>&1 | grep -i "x-litellm-model-id"

< x-litellm-model-id: 0d410bc56f642789e11dec790cf9c3ba46c680f1b6e25c5a9cf171bb1f2f3791

Success - Non-streaming - Cross-provider from OpenAI:

curl -v http://localhost:4000/v1/messages \
  -H 'Authorization: Bearer sk-master' \
  -H 'Content-Type: application/json' \
  -d '{"model": "gpt-4.1", "max_tokens": 10, "messages": [{"role": "user", "content": "Hi"}]}' \
  2>&1 | grep -i "x-litellm-model-id"

< x-litellm-model-id: 91f84a8c74066cd29acc087803ba323645417b95f9e55096bf8e3827149e8492

Success - Streaming - Cross-provider from OpenAI:

curl -v http://localhost:4000/v1/messages \
  -H 'Authorization: Bearer sk-master' \
  -H 'Content-Type: application/json' \
  -d '{"model": "gpt-4.1", "max_tokens": 10, "messages": [{"role": "user", "content": "Hi"}], "stream": true}' \
  2>&1 | grep -i "x-litellm-model-id"

< x-litellm-model-id: 91f84a8c74066cd29acc087803ba323645417b95f9e55096bf8e3827149e8492

Error - Bad Request - Cross-provider from OpenAI:

curl -v http://localhost:4000/v1/messages \
  -H 'Authorization: Bearer sk-master' \
  -H 'Content-Type: application/json' \
  -d '{"model": "gpt-4.1", "max_tokens": 10, "messages": [{"role": "user", "content": "Hi"}], "reasoning_effort": "low"}' \
  2>&1 | grep -i "x-litellm-model-id"

< x-litellm-model-id: 91f84a8c74066cd29acc087803ba323645417b95f9e55096bf8e3827149e8492

Error - Streaming Bad Request - Cross-provider from OpenAI:

curl -v http://localhost:4000/v1/messages \
  -H 'Authorization: Bearer sk-master' \
  -H 'Content-Type: application/json' \
  -d '{"model": "gpt-4.1", "max_tokens": 10, "messages": [{"role": "user", "content": "Hi"}], "reasoning_effort": "low", "stream": true}' \
  2>&1 | grep -i "x-litellm-model-id"

< x-litellm-model-id: 91f84a8c74066cd29acc087803ba323645417b95f9e55096bf8e3827149e8492

---

/v1/responses

Success - Non-streaming:

curl -v http://localhost:4000/v1/responses \
  -H 'Authorization: Bearer sk-master' \
  -H 'Content-Type: application/json' \
  -d '{"model": "gpt-4.1", "input": "Hi"}' \
  2>&1 | grep -i "x-litellm-model-id"

< x-litellm-model-id: 91f84a8c74066cd29acc087803ba323645417b95f9e55096bf8e3827149e8492

Success - Streaming:

curl -v http://localhost:4000/v1/responses \
  -H 'Authorization: Bearer sk-master' \
  -H 'Content-Type: application/json' \
  -d '{"model": "gpt-4.1", "input": "Hi", "stream": true}' \
  2>&1 | grep -i "x-litellm-model-id"

< x-litellm-model-id: 91f84a8c74066cd29acc087803ba323645417b95f9e55096bf8e3827149e8492

Error - Bad Request:

curl -v http://localhost:4000/v1/responses \
  -H 'Authorization: Bearer sk-master' \
  -H 'Content-Type: application/json' \
  -d '{"model": "gpt-4.1", "input": "Hi", "temperature": 2.5}' \
  2>&1 | grep -i "x-litellm-model-id"

< x-litellm-model-id: 91f84a8c74066cd29acc087803ba323645417b95f9e55096bf8e3827149e8492

Error - Streaming Bad Request:

curl -v http://localhost:4000/v1/responses \
  -H 'Authorization: Bearer sk-master' \
  -H 'Content-Type: application/json' \
  -d '{"model": "gpt-4.1", "input": "Hi", "temperature": 2.5, "stream": true}' \
  2>&1 | grep -i "x-litellm-model-id"

< x-litellm-model-id: 91f84a8c74066cd29acc087803ba323645417b95f9e55096bf8e3827149e8492

Success - Non-streaming - Cross-provider to Bedrock:

curl -v http://localhost:4000/v1/responses \
  -H 'Authorization: Bearer sk-master' \
  -H 'Content-Type: application/json' \
  -d '{"model": "claude-sonnet-4", "input": "Hi"}' \
  2>&1 | grep -i "x-litellm-model-id"

< x-litellm-model-id: 0d410bc56f642789e11dec790cf9c3ba46c680f1b6e25c5a9cf171bb1f2f3791

Success - Streaming - Cross-provider to Bedrock:

curl -v http://localhost:4000/v1/responses \
  -H 'Authorization: Bearer sk-master' \
  -H 'Content-Type: application/json' \
  -d '{"model": "claude-sonnet-4", "input": "Hi", "stream": true}' \
  2>&1 | grep -i "x-litellm-model-id"

< x-litellm-model-id: 0d410bc56f642789e11dec790cf9c3ba46c680f1b6e25c5a9cf171bb1f2f3791

Error - Bad Request - Cross-provider to Bedrock:

curl -v http://localhost:4000/v1/responses \
  -H 'Authorization: Bearer sk-master' \
  -H 'Content-Type: application/json' \
  -d '{"model": "claude-sonnet-4", "input": "Hi", "temperature": 2.5}' \
  2>&1 | grep -i "x-litellm-model-id"

< x-litellm-model-id: 0d410bc56f642789e11dec790cf9c3ba46c680f1b6e25c5a9cf171bb1f2f3791

Error - Streaming Bad Request - Cross-provider to Bedrock:

curl -v http://localhost:4000/v1/responses \
  -H 'Authorization: Bearer sk-master' \
  -H 'Content-Type: application/json' \
  -d '{"model": "claude-sonnet-4", "input": "Hi", "temperature": 2.5, "stream": true}' \
  2>&1 | grep -i "x-litellm-model-id"

< x-litellm-model-id: 0d410bc56f642789e11dec790cf9c3ba46c680f1b6e25c5a9cf171bb1f2f3791

[vercel]

@raghav-stripe is attempting to deploy a commit to the CLERKIEAI Team on Vercel.

A member of the Team first needs to authorize it.