Provider-level custom timeouts available on AI Gateway

AI Gateway now supports per-inference provider timeouts for fast failover. If a provider doesn't start responding within your configured timeout, AI Gateway aborts the request and falls back to the next available provider.

Provider timeouts are available in beta for BYOK (Bring Your Own Key) credentials only. Note that some providers don't support stream cancellation, so you may still be charged for timed-out requests depending on the provider.

Basic usage

Set timeouts per provider in milliseconds using providerTimeouts in providerOptions.gateway.

1
const result = streamText({
2
  model: 'openai/gpt-5.4',
3
  prompt,
4
  providerOptions: {
5
    gateway: {
6
      providerTimeouts: {
7
        byok: { openai: 15000 }, // 15 seconds
8
      },
9
    },
10
  },
11
});

Advanced usage with multiple providers and failover

Use with order to control both the provider sequence and failover speed.

1
const result = streamText({
2
  model: 'anthropic/claude-sonnet-4.6',
3
  prompt,
4
  providerOptions: {
5
    gateway: {
6
      order: ['anthropic', 'bedrock', 'vertex'],
7
      providerTimeouts: {
8
        byok: {
9
          anthropic: 10000,
10
          bedrock: 15000,
11
        },
12
      },
13
    },
14
  },
15
});

For more information, read the custom provider timeouts documentation.