The Vercel AI SDK with a custom base URL

Every model, chat completions

npm install ai @ai-sdk/openai-compatible

Claude models, Anthropic protocol

npm install ai @ai-sdk/anthropic

import { createOpenAI } from '@ai-sdk/openai';

const openai = createOpenAI({
  baseURL: 'https://aigrow.gg/v1',
  apiKey: 'ag_your_key_here',
});

const model = openai.chat('gpt-6-astra');

import { createAnthropic } from '@ai-sdk/anthropic';
import { generateText } from 'ai';

const anthropic = createAnthropic({
  baseURL: 'https://aigrow.gg/v1',
  apiKey: 'ag_your_key_here',
});

const { text } = await generateText({
  model: anthropic('claude-opus-5.5'),
  prompt: 'Reply with the word ready',
});

404 on /v1/responses

The model came from openai(...) in @ai-sdk/openai, which calls the Responses API. Use createOpenAICompatible as in the block, or openai.chat(...).

404 on the Anthropic provider

The baseURL lost its /v1. The Anthropic provider adds only /messages, so its baseURL is https://aigrow.gg/v1.

Requests still go to OpenAI

The ready-made openai import uses OPENAI_BASE_URL, or else api.openai.com. Create your own provider with the baseURL, as in the block.

My tool says 401 or authentication error

The key is missing or mistyped. Copy it again from API access on the dashboard, and make sure it is sent as the key, not as a model name. A revoked key fails the same way: create a new one and swap it in.

I get 402 or a message about credit

Your balance is empty. Open API access on the dashboard, redeem a code or buy another pack, and the next request goes through. Nothing else needs to change.

It says 404 or Unknown model

The model has to be one the endpoint serves, for example claude-opus-5.5. The provider’s own spellings work too (claude-opus-5-5, anthropic/claude-opus-5.5), but a dated id or any model outside the catalogue is rejected before the request is forwarded, and nothing is billed. Request GET /v1/models with your key for the live list, or read the ids on the price list.

Requests are rejected with 429

The endpoint’s capacity is shared by every buyer: about 20 requests a minute and 4 at once. Each 429 carries a retry-after header, and the OpenAI and Anthropic SDKs wait that long and retry on their own. In a loop of your own, wait the seconds it names or lower your concurrency.

The model is unavailable right now

A 404 that says the model is not available right now means it is briefly down at the upstream provider, not a problem with your key or balance. Retry in a moment or send the request to another model.

I get a 502 or 503

A transient upstream or configuration hiccup, not a charge against your balance. Retry the request; if it keeps failing, reach us through the store and we will look into it.

Vercel AI SDK is billed per token at the credit rates in the price list. An agent sends the whole conversation again with each step, so a long task costs more than its last message suggests; the activity log on API access shows what each request cost.

How do I set a custom base URL in the Vercel AI SDK?

Create a provider with createOpenAICompatible from @ai-sdk/openai-compatible, with baseURL set to https://aigrow.gg/v1 and apiKey set to your key, then pass it a model id such as claude-opus-5.5.

Why does @ai-sdk/openai give a 404 with a custom base URL?

Since AI SDK 5, openai(...) calls the Responses API at /v1/responses. The endpoint serves chat completions only, so create the model with openai.chat(...) or use the OpenAI Compatible provider.

Does the Anthropic provider take a custom baseURL?

Yes. createAnthropic accepts baseURL, and its default is https://api.anthropic.com/v1, so for AIGROW it is https://aigrow.gg/v1. Set apiKey or authToken to your key.

Do streamText and tool calls work?

Yes. streamText streams on both providers, and the tools you define with tool() are sent as function calls that your code runs. includeUsage: true adds token counts to a streamed result.

Sources: Vercel AI SDK OpenAI Compatible provider · Vercel AI SDK OpenAI provider · Vercel AI SDK Anthropic provider. Read 21 October 2026.