Gateway quickstart

Call any model through one endpoint with your Respan API key.

Paste this into your coding agent, or follow the setup below:


Help me set up the Respan gateway: fetch and follow https://www.respan.ai/docs/ai-gateway/quickstart.md

Setup

Create a key in Gateway › API keys , set it as RESPAN_API_KEY, and point your SDK at the gateway:

Terminal


curl https://gateway.respan.ai/chat/completions \
  -H "Authorization: Bearer $RESPAN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "gpt-5.5", "messages": [{"role": "user", "content": "Hello!"}]}'

Python


import os
from openai import OpenAI

client = OpenAI(
    base_url="https://gateway.respan.ai",
    api_key=os.environ["RESPAN_API_KEY"],
)

response = client.chat.completions.create(
    model="gpt-5.5",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)

TypeScript


import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://gateway.respan.ai",
  apiKey: process.env.RESPAN_API_KEY,
});

const response = await client.chat.completions.create({
  model: "gpt-5.5",
  messages: [{ role: "user", content: "Hello!" }],
});
console.log(response.choices[0].message.content);

Change model to any ID on respan.ai/models . Calls are paid from your Respan credits, or by your provider if you bring your own key.

Migrate from another provider or gateway

Most apps only change the base URL and the API key.

Keep the OpenAI SDK. Change the base URL and use your Respan API key. OpenAI model IDs stay the same, such as gpt-5.5.

Python


# Before
client = OpenAI(api_key=os.environ["OPENAI_API_KEY"])

# After
client = OpenAI(base_url="https://gateway.respan.ai", api_key=os.environ["RESPAN_API_KEY"])

To keep paying OpenAI directly, save your OpenAI key with BYOK.

Keep the Anthropic SDK. Change the base URL and use your Respan API key.

Python


# Before
client = anthropic.Anthropic(api_key=os.environ["ANTHROPIC_API_KEY"])

# After
client = anthropic.Anthropic(base_url="https://gateway.respan.ai", api_key=os.environ["RESPAN_API_KEY"])

For Claude Code, set ANTHROPIC_BASE_URL=https://gateway.respan.ai and ANTHROPIC_API_KEY to your Respan key, and unset ANTHROPIC_AUTH_TOKEN, which Claude Code prefers when both are set. See the Claude Code guide.

Change the base URL and key, then the model IDs. Move your fallback list into a fallback on the model.

Python


# Before
client = OpenAI(base_url="https://openrouter.ai/api/v1", api_key=os.environ["OPENROUTER_API_KEY"])
client.chat.completions.create(
    model="openai/gpt-5.5",
    messages=messages,
    extra_body={"models": ["anthropic/claude-sonnet-4.5"]},
)

# After: "smart" is a fallback route with entries gpt-5.5, then claude-sonnet-4-5-20250929
client = OpenAI(base_url="https://gateway.respan.ai", api_key=os.environ["RESPAN_API_KEY"])
client.chat.completions.create(model="smart", messages=messages)

Model IDs come from the model catalog and have no vendor prefix: gpt-5.5, not openai/gpt-5.5. To reach a model through OpenRouter itself, prefix it with openrouter/, as in openrouter/openai/gpt-oss-120b.

OpenRouter	Gateway
`models: [...]` fallback list	A fallback
`provider.order`	A fallback route whose entries are provider offerings, such as `azure/gpt-5.5`, then `gpt-5.5`
`provider.only`	A provider offering ID such as `azure/gpt-5.5`
`provider.allow_fallbacks: false`	A provider offering ID, which keeps the request on that provider
`user`	`customer_identifier`

Keep the AI SDK and your model IDs. Install @respan/ai-sdk-provider and make Respan the default provider once at startup. Each model is called in its native format, so options such as providerOptions.anthropic keep working.

Terminal


npm install @respan/ai-sdk-provider

TypeScript


// Before: string model IDs go to Vercel AI Gateway
const result = await generateText({ model: "anthropic/claude-sonnet-4.5", prompt });

// After: set once at startup, and the same IDs go to Respan
import { respan } from "@respan/ai-sdk-provider";

globalThis.AI_SDK_DEFAULT_PROVIDER = respan;

const result = await generateText({ model: "anthropic/claude-sonnet-4.5", prompt });

If you call gateway("...") directly, call respan("...") instead. openai/, anthropic/ and google/ IDs work as they are; for other models, use the ID from respan.ai/models . Vercel’s providerOptions.gateway.order maps to a provider offering ID or a fallback route of provider offerings, and its models fallback list to a fallback.

If you run the LiteLLM proxy, point your clients at the gateway instead. Your routing config moves to fallbacks and load balancers in the Respan platform.

Python


# Before
client = OpenAI(base_url="http://localhost:4000", api_key=os.environ["LITELLM_KEY"])
client.chat.completions.create(model="gpt-5.5", messages=messages, extra_body={"fallbacks": ["claude-sonnet-4-5-20250929"]})

# After: "smart" is a fallback route with entries gpt-5.5, then claude-sonnet-4-5-20250929
client = OpenAI(base_url="https://gateway.respan.ai", api_key=os.environ["RESPAN_API_KEY"])
client.chat.completions.create(model="smart", messages=messages)

LiteLLM model_list entries with weights become a load balancer across models, or key weights across your saved provider keys. If you call the LiteLLM SDK in-process, see the LiteLLM guide.

Logs

Every call shows up in Logs with its model, provider, tokens, latency, and cost.

Using a framework or coding agent? See Integrations.