Recraft V3 SVG
Recraft's vector/SVG generation model. Editable illustrations and icons from text.
Recraft V3 SVG is image generation AI model from Recraft, priced at €0.000 per 1M input tokens with a unknown context window.
Pricing
API Integration
Use our OpenAI-compatible API to integrate Recraft V3 SVG into your application.
npm install railwailimport railwail from "railwail";
const rw = railwail("YOUR_API_KEY");
const images = await rw.run("recraft-v3-svg", "A beautiful sunset over Tokyo");
console.log(images[0].url);
// Or use the image() method for full control
const res = await rw.image("recraft-v3-svg", "A cat in space", {
size: "1024x1024",
n: 1,
});
console.log(res.data[0].url);Deep dive — Recraft AI's Recraft V3 SVG
Recraft AI is a generative-AI startup founded in 2022 in London by Anna Veronika Dorogush (CEO, former Yandex ML lead and creator of the CatBoost gradient-boosting library) and other ex-Yandex engineers. The company raised a $12M Series A led by Khosla Ventures in 2024 and a follow-on round in early 2025, focusing on AI tools for designers, brand teams and marketing agencies. Recraft's main product is a design-focused web app and its proprietary Recraft V3 model (released October 2024), which topped the Artificial Analysis Image Arena leaderboard at launch ahead of Midjourney v6.1, FLUX 1.1 [pro], DALL-E 3 and Ideogram 2.0 on human preference and prompt adherence. Recraft is positioned as a commercial design platform with strong typography, brand-style consistency and vector-output capabilities. Recraft V3 SVG is the variant specialised for native vector graphics output.
Visit Recraft AI →Recraft V3 SVG is a closed-source diffusion image model with a custom vector-output decoder that emits scalable vector graphics directly rather than rasterising them from pixels and post-processing. The underlying backbone is the same Recraft V3 model (codenamed 'red panda' during pre-release benchmarking) that topped the Artificial Analysis text-to-image arena in October 2024 at Elo ~1172 — widely reported as a Diffusion Transformer with a long-context LLM-based text encoder handling ~1000-token prompts. The SVG variant adds a vector decoder trained to produce clean, editable SVG paths and shapes suitable for direct use in Figma, Adobe Illustrator, Sketch and Inkscape. Recraft has not published a technical report or disclosed parameter counts. The SVG output handles logos, icons, illustrations and infographics; very intricate photorealistic scenes are not ideal SVG candidates. Available via the Recraft web app and API.
- Parameters
- Undisclosed
- Context
- 1.0K tokens
- Native SVG output — directly editable in Figma, Illustrator, Inkscape
- Built on Recraft V3 backbone (#1 on Artificial Analysis Image Arena at launch)
- Industry-leading in-image text rendering at release
- Long-prompt understanding (~1000 tokens)
- Brand-style consistency via Recraft Styles libraries
- Layout-aware generation: logos, icons, posters, infographics
- Distinct from raster-to-vector post-processing — emits SVG natively
- Best for: logos, icons, infographics, scalable design assets, brand-consistent illustrations.
Trained on a large licensed and publicly available image-text dataset with heavy design-focused augmentation. Recraft has publicly stated it licenses or commissions design data rather than relying on the scale of web-scraped general image models. SVG decoder training adds curated vector-graphic data. Exact composition not disclosed.
License: Recraft Terms of Service (closed hosted API). Standard image-generation commercial terms; paying subscribers retain rights to generated assets for commercial use under Recraft's TOS.
Known limitations
- Closed weights — hosted API only, no self-hosting
- SVG output complexity is bounded — very intricate scenes do not vectorise cleanly
- Less photorealistic than FLUX.1 Pro or Midjourney v6 for natural-scene generation
- Pricing higher than commodity image models, reflecting design-pro positioning
- No video / no 3D output
- No public weights or fine-tuning
Frequently asked questions
Related Models
View all Image GenerationFlux 1.1 Pro Ultra
FLUX 1.1 Pro in ultra mode. Up to 4 megapixel images with raw mode for photorealism.
Flux Dev
Black Forest Labs' development model. Fast, high-quality image generation with LoRA support.
Google Imagen 4
Google's Imagen 4. Text-to-image with strong photorealism and improved typography support.
Google Imagen 4 Ultra
Premium Imagen 4 tier. Highest fidelity, prompt adherence and typography quality from Google.
Start using Recraft V3 SVG today
Get started with free credits. No credit card required. Access Recraft V3 SVG and 100+ other models through a single API.