whatmodel.sh / image / Krea 2 Turbo
Krea 2 Turbo
↗ share
+ contribute
Krea 2 Turbo is an open-weights image model from Krea, released June 2026. Its published size is 12B parameters. Licence: krea-2-community-license.
Catalogue identity notes
Comparison-only. Model card (krea/Krea-2-Turbo @98e0fe1): Release Date June 22, 2026; Diffusion Transformer with 12 billion parameters; krea-2-community-license (auto-approved gate). Arena entry 'krea-2-turbo' (krea, krea-2-community-license).
Score within the LMArena text-to-image comparison set only.
$ how it compares
The image models ranked either side of Krea 2 Turbo, out of 197 with a score, then more from Krea.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 54 |  Recraft V4 | 59±3 | api $0.040/image | ||
| 55 |  FireRed-Image-Edit-1.1 | 59±4 | comparison only | ||
| 56 | Krea 2 Turbo | 59±3 | api $0.015/image +local | ||
| 57 |  GPT Image 1 Mini | 59±2 | api $0.011/image | ||
| 58 | Krea 2 Large | 59±3 | api $0.060/image | ||
| more from Krea | |||||
| 51 | Krea 2 Medium | 60±3 | api $0.030/image | ||
$ what it's good at
text to image▂▅▅▃▄▄█▃▁▂▁▁ #46 of 126 * · 97% of its score · 1 test
Beats Wan 2.7 Image here, though they rank higher overall. Trails FLUX.2 [dev], GPT Image 1 Mini here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| LMArena text-to-image | 1113.2 ✎ | ▂▂▃▄▇█▄▂▄▂·▂ beats 29/75 | 97% |
$ where to run it
Through an API
api · $0.015/image · www.krea.ai ↗ · price sourceUSD per generated image, Krea 2 Turbo text-to-image on the Krea API: fixed price $0.015 (style references $0.0175, moodboards $0.02). Product: Krea API. API model: Krea 2 Turbo (POST /generate/image/krea/krea-2/medium-turbo). Settings: resolution=1K (the only option), no references, one image. Price basis: published fixed list price. Identity: the open-weights card (krea/Krea-2-Turbo) describes Turbo as the post-trained, distilled release checkpoint with 'Open-weight release and Krea-hosted product integrations'; fal also serves the open weights (fal-ai/krea-2/turbo) at $0.008 per megapixel. Currency: USD. Krea official API documentation, observed 2026-09-25.
Other ways to run it
version
- Run it as published · needs ~20.2 GB · from github.com/leejet
sd-cli --diffusion-model models/Krea-2-Turbo-Q8_0.gguf --llm models/Qwen3VL-4B-Instruct-Q4_K_M.gguf --vae models/wan_2.1_vae.safetensors -p "a lovely cat holding a sign says 'krea2.cpp'" --steps 8 --cfg-scale 1.0 --diffusion-fa -v --offload-to-cpu
Official weights and documentation: huggingface.co/krea ↗
$ 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
| source | results | share of the score |
|---|---|---|
 LMArena leaderboard dataset | 1 | 97% |
| prior · pull toward the middle | 3% | |
| weighted average | = 58.8 |
$ improve this page
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