Silent model
Retired Grok slugs still resolve but redirect to other models, so the model served and the bill change silently; grok-code-fast-1's target, grok-build-0.1, reached end of life November 19, 2026
What changes
From May 15, 2026 12:00 PM PT, requests to the retired slugs redirect instead of erroring. grok-4-1-fast-reasoning, grok-4-fast-reasoning and grok-4-0709 go to grok-4.3 at low reasoning effort; grok-4-1-fast-non-reasoning, grok-4-fast-non-reasoning and grok-3 go to grok-4.3 at none; grok-code-fast-1 goes to grok-build-0.1, which itself reached end of life on November 19, 2026 and is no longer served (xAI does not say what grok-code-fast-1 requests do since then); grok-imagine-image-pro goes to grok-imagine-image-quality, which is itself retired November 2, 2026. xAI says requests to a retired slug after May 15 are billed at grok-4.3 pricing ($1.25 per 1M input, $2.50 per 1M output).
Silent: No error. Behavior, results, or cost change quietly.
What to do
Set the model explicitly instead of relying on the redirect: grok-4.3 with an explicit reasoning_effort, grok-4.7 for code (xAI's Models page recommends Grok 4.7 for code; grok-build-0.1, the grok-code-fast-1 redirect target, reached end of life November 19, 2026), and grok-imagine-image-2.0 for grok-imagine-image-pro (its redirect target, grok-imagine-image-quality, retired November 2, 2026).
Strings to search for
grok-4-1-fast-reasoninggrok-4-1-fast-non-reasoninggrok-4-fast-reasoninggrok-4-fast-non-reasoninggrok-4-0709grok-code-fast-1grok-3grok-imagine-image-pro