
Recraft V4
#1 on the HF text-to-image arena with native editable SVG vector output.
About
Recraft V4 is a ground-up rebuild of Recraft's image model, released February 17, 2026, designed with professional designers rather than just trained on datasets. It topped the Hugging Face Text-to-Image Arena with an ELO of 1172 and a 72% win rate against Midjourney, DALL-E 3, and FLUX. Its killer feature is native SVG vector output — the only generative model that produces editable, layered vector files directly from a text prompt, alongside standard raster generation (1024×1024, Pro at 2048×2048). Recraft's positioning centers on 'design taste': composition, lighting, color relationships, and precise text rendering as first-class concerns.
The honest picture: V4 was superseded by V4.1 in May 2026 (sharper photorealism, better prompt following), though V4 remains available via API. It's a proprietary API-only model — no open weights, priced per image (~$0.04 raster / $0.08 vector). The SVG output still needs professional review for geometry and brand compliance; 'editable' doesn't mean production-ready. And Recraft is unambiguously a design tool, not a general image model — for photorealistic people and scenes, dedicated photo models still win.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.
Use cases
Trust score — 8/10
Scored 8/10 (verified tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 47; Trending score 80/100 from researcher assessment.