whatmodel.sh / image / Recraft V4

Recraft V4

+ contribute
Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Recraft V4 is a closed-weights image model from Recraft, released February 2026. Licence: proprietary.

58.9±2.5#47 of 144 image modelsfrom 1 benchmark result across 1 source* thinly tested, treat the rank as provisional

Score within the LMArena text-to-image comparison set only.

$ how it compares

The image models ranked either side of Recraft V4, out of 197 with a score, then more from Recraft.

#modelscorewhereRAMreleased
52Imagen 4 Google59±2api $0.040/image2025-05
53Wan 2.5 Text-to-Image Preview Alibaba59±3api $0.030/image2025-09
54Recraft V4 Recraft59±3api $0.040/image2026-02
55FireRed-Image-Edit-1.1 FireRedTeam (Xiaohongshu)59±4comparison only2026-03
56Krea 2 Turbo Krea59±3api $0.015/image +local~20GB2026-06
more from Recraft
31Recraft V4.1 Utility Pro Recraft64±2api $0.21/image2026-05
46Recraft V4.1 Pro Recraft61±3api $0.21/image2026-05
67Recraft V4.1 Flash Recraft57±3api $0.0070/image2026-09

see all image models · find the ones that fit your machine

$ what it's good at

text to image▂▅▅▃▄▄█▃▁▂▁▁ #44 of 126 * · 97% of its score · 1 test

Beats Wan 2.7 Image here, though they rank higher overall. Trails FLUX.2 [dev] here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
LMArena text-to-image1114.9 ✎▂▂▃▄▇█▄▂▄▂·▂ beats 30/75avg 1027▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 97%typical

$ where to run it

Through an API

api · $0.040/image · www.recraft.ai ↗ · price sourceUSD per generated raster image, Recraft V4 (standard; V4 Pro $0.25, vector output priced separately). Billed as 40 API units per image from pre-purchased unit packages. Product: Recraft API raster image generation. API model: Recraft V4. Settings: raster, n=1. Price basis: published list rate. Currency: USD. Recraft official API pricing page, observed 2026-09-25.

$ how the score was worked out · LMArena leaderboard dataset 97%

Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works

sourceresultsshare of the scorelicence
LMArena leaderboard dataset1▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 97%CC BY 4.0
prior · pull toward the middle▇ 3%
weighted average= 58.9

$ improve this page

Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.

data updated · sources and licences · json