grok-imagine-image-quality retires Nov 2: what about quality?
xAI retires grok-imagine-image-quality on November 2, 2026 and serves it as grok-imagine-image-2.0 at low quality. Sume's x-ai/grok-image takes no quality.

On November 2, 2026, xAI retires the grok-imagine-image-quality slug and serves requests to it with grok-imagine-image-2.0 at quality: low, with no change to the request or response shape. On Sume there is no quality slug to migrate: Grok Imagine is the single id x-ai/grok-image, and it advertises no quality parameter.
Vendor facts are from xAI's release notes and image guide; Sume facts from the Image API docs and catalog code. All read 2026-09-30.
What did xAI change?
Two notes matter. The September 2 note sets the retirement. The August 28 note says quality on grok-imagine-image-2.0 now accepts auto, that omitting it now means auto instead of medium, and that images are billed at the quality they are served at.
| Item | As written by xAI |
|---|---|
grok-imagine-image-quality | Retired November 2, 2026; served by grok-imagine-image-2.0 with quality set to low |
grok-imagine-image (1.0) | Not affected |
quality values on 2.0 | low, medium, auto (the default when omitted) |
What auto picks | low for generation, medium for editing, currently |
| Billing | At the quality served |
Does Sume's Grok id accept quality?
No. The router only publishes a quality descriptor when the catalog lists qualities for that model, and its public quality table covers GPT Image and Ideogram, not Grok. The docs say a parameter the model does not list is rejected with 400 unsupported_parameter instead of being dropped, so a request that carries quality to x-ai/grok-image fails rather than silently picking a tier. See the 400 rule.
What should I change in my requests?
If you call xAI directly with the retiring slug, switch the model to grok-imagine-image-2.0 and set quality yourself; low matches what the slug will become. If you call Sume, remove quality from the body and read the endpoint's supported_parameters for what the id accepts. The Sume id, ratios and edit limits are in Grok Imagine on Sume.
How is a Sume Grok image billed?
Per image, as one flat price line, and only when the generation completes; the docs say failed or cancelled generations are not billed. Read the current figure from the catalog entry rather than from a post.
Sources
Related posts
More in Models
- Grok Imagine video editing: 8.7 s on xAI, Omni edit on Sume
xAI caps Grok Imagine video edits at 8.7 seconds. Sume's Grok row has no video-to-video; edits go through gemini-omni-flash-1.1 with a video_url.
- Grok Imagine video sound: xAI audio vs Sume's silent row
xAI's Grok Imagine video makes audio unless generate_audio is False. Sume's grok-imagine-video-1.5 row lists audio: false and rejects generate_audio.
- HeyGen ElevenLabs v3 model_id per request vs Sume TTS
HeyGen's speech endpoint takes settings.model_id such as eleven_v3 per request. Sume TTS 1.0 hides model ids; its separate TTS Router lists models explicitly.
- HeyGen photo avatar hand gestures vs Sume Motion Control
HeyGen's motion_prompt still needs an animation reference for Avatar V photo avatars. On Sume, body motion is a separate Motion Control route, not a prompt.
Written by Sume