Grok Imagine Image 2.0 Climbs 14 Spots to #4 on Text-to-Image Leaderboard, Best Non-OpenAI Model
ArtificialAnlys · x · 2026-09-19
Artificial Analysis' latest leaderboard places xAI's Grok Imagine Image 2.0 at #4 on text-to-image, the highest-ranked model outside OpenAI and a 14-place climb over the previous generation, landing on the quality-vs-price Pareto frontier.
Released in August, the model supports text-to-image, image editing, and multi-reference generation with up to five input images. It jumped from #18 to #4 in text-to-image and #16 to #10 in image editing, with the largest gains in Knowledge, Text Rendering, and Lighting. It's available via the xAI API as grok-imagine-image-2.0 and in Grok Imagine.
More from Multimodal
- Gemini TTS tip: specifying accents in the prompt unlocks far better non-English performance — juberti · 2026-09-19
- LTX 2.5 CQ Enhancer LoRA Delivers Striking Upscale Results — blastbottles · 2026-09-19
- Modal x Gray Area picks 13 AI art commissions, each with $20K cloud credits — charles_irl · 2026-09-19
- AI Short Film 'Straw Tears' Reveals Full Pipeline: Midjourney, Seedance 2, Magnific — LudovicCreator · 2026-09-19
- Creator demos Higgsfield Genjutsu video gen with a friend's avatar — chrisfirst · 2026-09-19
- GTA VI trailer parody built with MiniMax H3, custom LoRAs, Yue2 music and CapCut — aziib · 2026-09-19