Grok Imagine Image 2.0
SpaceXSI's (formerly xAI) second-generation Grok Imagine image model for generating and editing images from text or image prompts. It sits in the tied second group on LMArena text-to-image and image editing.
SpaceXSI (formerly xAI)Image modelReleased August 2026Data checked October 18, 2026
Where people rank it
#2
Editing images (tied), 50,427 votes
#3
Creating images (tied), 14,483 votes
From blind votes on LMArena, where people pick the better of two anonymous answers. See the full tables on the AI leaderboard.
Strengths and weaknesses
- Tied for second on LMArena text-to-image and image editing
- Low price, from $0.04 per image on the API
- Accepts both text and image prompts
- Its best LMArena score comes from a 'canvas' variant with few votes; the dated API snapshot scores lower (1294.2)
- Well behind GPT-Image-2.5 Sunburst on every image arena
Best for
- Fast, cheap image generation
- Image edits from a reference photo
Every score we have
Each bar compares this model with the best result any model on this page has on the same test.
- LMArena text to image1,333.5Arena score 1333.5 (95% CI 1322.0 to 1345.1); 3,113 votes; entry "grok-imagine-image-2.0 (canvas)"; position 4 of 82 on the full board
- LMArena image editing1,439.1Arena score 1439.1 (95% CI 1430.5 to 1447.7); 11,589 votes; entry "grok-imagine-image-2.0 (canvas)"; position 4 of 59 on the full board
- Artificial Analysis image arena1,166.8Elo 1166.8 (95% CI 1153.8 to 1179.8); 7,245 appearances; position 5 of 104 on the Artificial Analysis text-to-image arena
- Price per 1,000 images (estimate)$60$60.00 per 1,000 images (Artificial Analysis estimate)
Compare with
- MAI-Image-2.6Microsoft AI's image model, launched on August 10, 2026 and offered in public preview on Microsoft Foundry.
- Muse ImageMeta's agentic image model, which plans and checks its work before rendering and can ground images in search results.
- Seedream 5.0 ProByteDance Seed's image creation and editing model, launched on July 8, 2026, with region editing, layer splitting and dense infographic text.
- GPT-Image-2.5 SunburstOpenAI's most capable image generation and editing model, released as a September 8, 2026 snapshot alongside the faster GPT-Image-2.5 Flare.