# Recraft V3

> Legacy. Recommend recraft-v4 for new projects (same price, top of HF Arena).

Human version: https://kymaapi.com/models/recraft-v3
Live JSON: `GET https://kymaapi.com/v1/models` (no auth required)

## Facts

- **Model ID**: `recraft-v3` — pass this as `model` in the request body
- **Creator**: Recraft
- **Released**: 2024-10-30
- **Price**: $0.054 / image

## Call it

```bash
curl https://kymaapi.com/v1/images/generations \
  -H "Authorization: Bearer $KYMA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "recraft-v3", "prompt": "..."}'
```

## Positioning

Recraft V3 is a legacy text-to-image model optimized for generating vectors, illustrations, and brand assets. It suits developers maintaining existing pipelines, though new projects should target its successor.

## About

Recraft V3 is a text-to-image model from Recraft that specializes in vector graphics, illustrations, and brand-ready assets. It operates on a per-image pricing model and is classified as a strong-quality, cheap-tier generator with medium inference speed.

On Kyma, the model accepts text prompts and returns images through an OpenAI-compatible endpoint. It supports prompt caching, and every request benefits from automatic failover if a serving path degrades. Responses include exact cost in usage.cost and report the active model via the X-Kyma-Model header.

This model does not support vision, reasoning, or structured outputs, and it lacks a configurable context or output token window. As a legacy release, it is maintained for backward compatibility rather than active development.

## Use cases

- **Vector graphic generation** — Create scalable SVG-style assets directly from text prompts.
- **Brand asset creation** — Produce consistent illustrations and logos for marketing materials.
- **Legacy pipeline support** — Maintain existing image generation workflows without migrating endpoints.
- **Rapid visual prototyping** — Generate quick visual mockups on a cheap tier.

## Not ideal for

Do not use this model for photorealistic generation, complex multi-step reasoning, or any workflow requiring vision input or structured JSON outputs.

## Pick something else when

- You need the latest Recraft architecture for new projects → `recraft-v4`
- You require high-fidelity photorealistic generation → `flux-2-pro`, `imagen-4-ultra`
- You need vector output with advanced editing capabilities → `recraft-v4-vector`
- You want faster inference speeds for rapid iteration → `imagen-4-fast`

## See also

- All models: https://kymaapi.com/models.md
- Pricing: https://kymaapi.com/pricing.md
- Other models by Recraft: https://kymaapi.com/models?q=Recraft