Recraft Models by Recraft
RecraftProprietary
2 models2024–2025
Last refreshed 2026-10-17. Next refresh: weekly.
Capabilities
MultimodalAll models
Links
WebsiteAbout
Recraft AI's image generation family, specializing in high-quality vector and raster image generation with design-focused features and brand consistency controls. Recraft V3 won the LMSYS text-to-image arena in October 2024.
Decision facts
- Best fit
- image generationvision and multimodal work
- Capability starting point
- Recraft V4 with multimodal inputs
- Lowest tracked input
- Not tracked
- Closest related family
- 360Zhinao
Current Variants
Use-when guidance is based on each model's tracked capabilities, context window, release date, and replacement status.
1 in view1 retired
Recraft V4Current
Use when the workload needs image generation and multimodal inputs.
2025-02image generationmultimodal inputs
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| Recraft V4 | Use when the workload needs image generation and multimodal inputs. | 2025-02 | image generationmultimodal inputs | Current |
Release Timeline
2 release groups2025-02
1 current
Recraft V4
Currentimage generationmultimodal inputs
2024-10
1 retired
Recraft V3
Replacedimage generationmultimodal inputs
Replaced By
Replaced
Keep for legacy integrations; evaluate Recraft V4 before new work.
Specifications(2 models)
| Model | Released | Multimodal |
|---|---|---|
| Recraft V4 | 2025-02 | Yes |