modelrelay_
Sign in

Your first request

Every example below sends the same prompt. They assume a secret key in MODELRELAY_API_KEY; see Getting started to create one. Use any model ID from the models page or mrl model list.

Responses API

POST /api/v1/responses is ModelRelay’s provider-neutral API. The same request shape works for every model. Send the key in X-ModelRelay-Api-Key; this endpoint rejects secret keys sent as Authorization: Bearer.

curl https://api.modelrelay.ai/api/v1/responses \
  -H "X-ModelRelay-Api-Key: $MODELRELAY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-5-5",
    "input": [
      {"type": "message", "role": "user", "content": [{"type": "text", "text": "What is the capital of France?"}]}
    ]
  }'

The response contains output items, usage, the model, and a stop reason. See the Responses API reference for the full schema.

OpenAI SDK

The /v1 endpoints accept the OpenAI wire format. Set the base URL to https://api.modelrelay.ai/v1 and pass your ModelRelay key as the API key.

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ["MODELRELAY_API_KEY"],
    base_url="https://api.modelrelay.ai/v1",
)

completion = client.chat.completions.create(
    model="claude-sonnet-5-5",
    messages=[{"role": "user", "content": "What is the capital of France?"}],
)
print(completion.choices[0].message.content)

client.responses.create(...) works against POST /v1/responses with the same client. Without an SDK:

curl https://api.modelrelay.ai/v1/chat/completions \
  -H "Authorization: Bearer $MODELRELAY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-5-5",
    "messages": [{"role": "user", "content": "What is the capital of France?"}]
  }'

See Chat Completions compatibility for supported fields.

Anthropic SDK

POST /v1/messages accepts the Anthropic Messages format. The Anthropic SDK adds /v1/messages to its base URL, so set the base URL to https://api.modelrelay.ai.

import os
import anthropic

client = anthropic.Anthropic(
    api_key=os.environ["MODELRELAY_API_KEY"],
    base_url="https://api.modelrelay.ai",
)

message = client.messages.create(
    model="claude-sonnet-5-5",
    max_tokens=1024,
    messages=[{"role": "user", "content": "What is the capital of France?"}],
)
print(message.content[0].text)

Without an SDK:

curl https://api.modelrelay.ai/v1/messages \
  -H "x-api-key: $MODELRELAY_API_KEY" \
  -H "anthropic-version: 2023-06-01" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-5-5",
    "max_tokens": 1024,
    "messages": [{"role": "user", "content": "What is the capital of France?"}]
  }'

See Messages compatibility for supported fields.

mrl

mrl "What is the capital of France?" --model claude-sonnet-5-5

Add --stream to print output as it arrives and --usage to print token usage. Piped input is prepended to the prompt:

git diff | mrl "Explain these changes" --model claude-sonnet-5-5

Set a default model with mrl config set --model claude-sonnet-5-5 to omit --model. See the CLI reference.

Next steps