This Flux LoRA generates scenes with two distinct characters: a red-haired heroine and a green zombie. It maintains separate features to avoid visual bleeding.
Models
All Models
This Flux LoRA generates a retro hero boy in neon outfits. The style reflects 80s arcade gaming with digital helmets, glowing visors, and futuristic elements.
This Flux LoRA generates modular RPG assets and strategy game props. It produces stylized stone arches, market stalls, and greenery for building environments.
This Flux LoRA generates vibrant, cartoonish battlegrounds for RPGs. It produces fantasy environments like volcanic pits and forests with bold, clean details.
This Flux LoRA produces fantasy creatures in a modern animated style. It features smooth brushstrokes, soft shading, and dynamic lighting on black backgrounds.
This Flux LoRA produces glossy, 3D icons with rounded shapes. It creates vibrant UI elements and characters with a polished, soft style and bold proportions.
This Flux LoRA generates whimsical architectural structures in isometric views. It produces detailed fantasy buildings with clean lines and bright colors.
This Flux LoRA produces clean, modern 3D assets with smooth surfaces and rounded edges. It creates minimalist props and vehicles for mobile apps or games.
This Flux LoRA creates anthropomorphized objects with expressive faces. It applies a vibrant cartoon style to furniture, food, and tools using wide eyes.
Luma Video Reframe (May 2025) adjusts the aspect ratios of a video by outpainting beyond the original frame while maintaining visual coherence and scene logic
Dreamina 3.1 by ByteDance generates images with enhanced aesthetics and diverse artistic styles. Pricing from 10 CU.
deogram Character by Ideogram is a specialized model for maintaining character consistency from a single reference image.
Qwen Image by Alibaba is a foundational text-to-image model with strong text rendering and artistic style support. Generation from 4 CU.
A model to upscale skyboxes while preserving the seamless aspect
A model to upscale textures & materials while preserving their seamless aspect
Flux-based upscaling model by Scenario
Aleph by Runway is a context-aware video editing model capable of scene reconstruction and novel camera angle generation.
Wan 2.2 T2V is an open-source text-to-video model by Alibaba with a Mixture-of-Experts architecture and Last Frame conditioning. From 15 CU.
FLUX 1 Krea (Dev): An "opinionated" model emphasizing natural textures and diverse aesthetics to avoid the "AI look." Pricing from 4 CU.
Wan 2.2 I2V is an open-source image-to-video model by Alibaba with a Mixture-of-Experts architecture and Last Frame conditioning. From 35 CU.
Runway Upscale V1 by Runway enhances generated video resolution to 4K while maintaining temporal consistency across frames.
Runway Gen4 Turbo by Runway is an early 2025 high-speed image-to-video generation model optimized for low-latency iteration. First frame is required. 36 CU
PartCrafter by PKU: A compositional DiT model generating up to 16 semantically distinct, separate 3D meshes from a single RGB image.
MiniMax Video 02 (Hailuo 02) by MiniMax produces realistic video with natural human movement. Pricing from 72 credits per generation.
MiniMax Image-01 (Apr 2025) specializes in complex lighting effects like sub-surface scattering for highly realistic skin.
Google's Veo 3 (Fast) is a speed-optimized variant of Veo 3, enabling rapid iteration from text or images with a quality and cost trade-off.
Recraft v3 SVG by Recraft generates scalable vector graphics for logos and icons. Pricing from 9 credits.
Recraft Crisp Upscale increases overall quality, making visuals suitable for web use or print-ready materials
Recraft Creative upscales image with a creative style.
Use this fast version of Imagen 4 when speed and cost are more important than quality.
Imagen 4 (Ultra) by Google delivers the highest level of detail and prompt adherence for professional creative work. Generation in about 6 seconds, from 10 credits.
Luma Photon by Luma AI generates high-fidelity images with professional prompt adherence starting from 4 credits.
Luma Photon Flash by Luma AI provides accelerated image generation for rapid prototyping and concept iteration from 2 credits.
Ideogram 3 (Quality) by Ideogram specializes in professional-grade typography and complex layout generation. Pricing from 10 credits.
Ideogram 3 (Balanced) by Ideogram offers an optimal mix of visual fidelity and generation efficiency. Pricing from 7 credits.
Imagen 3 (Fast) by Google generates images with accelerated speed for rapid iteration and high-volume tasks. Generation in about 3 seconds, from 5 credits.
Ideogram 3 (Turbo) by Ideogram provides rapid image generation for high-volume creative exploration. Pricing from 4 credits.
Seedream 3 by ByteDance generates native 2K images with fast processing and bilingual text rendering. Pricing from 4 CU.
Tripo 2.5 Multiview is a spatial consistency model reconciling 2–4 images for symmetric and dimensionally accurate 3D assets.
Rodin HighPack is a high-fidelity generation mode for Rodin Gen-1 supporting 4K textures and dense 500k geometry.
Rodin Gen-1 is an image-to-3D model with multi-view support for reconstructing consistent geometry and UVs.
Kling 2.1 (Master) by Kuaishou generates 1080p video with advanced 3D motion and refined facial modeling for cinematic quality. Pricing from 210 CU.
Hunyuan 3D 2.1 is a legacy single-image generator with photorealistic PBR material synthesis and refined character fidelity.