Fast Models
| Model | Description | Best For | Speed | Credits |
|---|---|---|---|---|
| Krea 1 | Most creative model with LoRAs. Excels at artistic photorealism. Realistic textures. | Artistic photorealism, LoRA styles | 3/3 | 6 |
| Flux 2 Klein | A lightweight 4 billion parameter version of Flux 2, optimized for speed while maintaining quality. Supports up to 2K resolution. | |||
| Qwen 2512 | Latest Qwen model with enhanced human realism, finer natural detail for landscapes and textures, and improved text rendering with better layout and composition. | Human portraits, landscapes, text rendering | 2/3 | 9 |
| Flux | Best model for LoRAs. Fastest model. Highly customizable, less realistic colors and textures. | LoRA customization, fast iteration | 3/3 | 5 |
| Z Image | Cheapest model. Medium quality photorealism at a budget. Realistic textures. | Budget photorealism, quick drafts | 3/3 | 3 |
Intelligent Models
| Model | Description | Best For | Speed | Credits |
|---|---|---|---|---|
| Nano Banana | Smart model optimized for precise prompt adherence. Best for most tasks. | General use, precise prompt following | 3/3 | ~30 |
| Nano Banana Pro | Smartest model. World’s best prompt adherence. Best model for complex tasks and image editing. | Complex tasks, image editing | 2/3 | ~100 |
| Nano Banana 2 | Google’s latest flash image model (also known as Gemini 3.1 Flash Image) optimized for fast generation with support for up to 4K resolution. | Fast high-res generation, 4K output | 2/3 | ~50 |
| Seedream 5 Lite | Latest high-quality model from ByteDance with deep thinking, web search capabilities, and up to 14 style images. | Deep thinking, multi-style generation | 2/3 | ~30 |
| ChatGPT 1.5 | Smartest model. Excellent prompt following. Best for complex diagrams, characters, and image editing. | Complex diagrams, characters, editing | 1/3 | ~150 |
| ChatGPT Image | Older ChatGPT image model. Intelligent prompt adherence with sharp semi-realistic output. | Semi-realistic output, prompt adherence | 1/3 | ~200 |
Quality Models
| Model | Description | Best For | Speed | Credits |
|---|---|---|---|---|
| Recraft V4 | Sharp, detailed images from Recraft with Standard and Pro modes. | Sharp detail, graphic design | 2/3 | ~30 |
| Seedream 4 | Latest high-quality model from ByteDance with flexible resolution support. | Flexible resolution, quality output | 2/3 | 25 |
| Seedream 4.5 | Quality 4K model optimized for vibrant photorealism. | Vibrant photorealism, 4K | 2/3 | ~30 |
| Flux 2 | FLUX.2 [dev] from Black Forest Labs. Enhanced realism and crisper text generation. | Realism, text generation | 3/3 | 20 |
| Flux 2 Pro | Medium quality model by Black Forest with stable visuals. | Stable, consistent visuals | 2/3 | ~60 |
| Flux 2 Flex | BFL’s next generation model, excelling at rendering text and fine details. | Text rendering, fine detail | 2/3 | ~200 |
| Flux 2 Max | Frontier model by Black Forest Labs. Most capable Flux 2 model. | Highest quality Flux output | 2/3 | ~400 |
| Wan 2.2 | Cinematic outputs with crisp textures and realistic color palette, but weak diversity and structure. | Cinematic, crisp textures | 1/3 | ~30 |
| Qwen | Semi-realistic model with great text generation and prompt adherence. Capable of following long and highly detailed prompts. | Long detailed prompts, text generation | 2/3 | 9 |
| Kling O1 | Kling Omni image model with support for up to 10 reference images with tags. | Multi-reference image generation | 2/3 | 20 |
4. Legacy Models
| Model | Description | Best For | Speed | Credits |
|---|---|---|---|---|
| Flux Kontext | Frontier model designed for image editing. Capable of advanced reasoning and style transfer. | Image editing, style transfer | 3/3 | 9 |
| Flux Kontext Pro | Frontier model designed for image editing. Capable of advanced reasoning and style transfer. | Advanced image editing, style transfer | 2/3 | ~30 |
| Imagen 4 Ultra | Google’s best image model. Excellent prompt adherence. Capable of producing a wide range of subjects. | Wide subject range, high quality | 2/3 | ~45 |
| Imagen 4 Fast | Google’s fastest image model. Excellent prompt adherence. Capable of producing a wide range of subjects. | Fast Google-quality output | 2/3 | 15 |
| Imagen 4 | Google’s best image model. Excellent prompt adherence. Capable of producing a wide range of subjects. | General high-quality generation | 1/3 | ~30 |
| Runway Gen-4 | Cinematic image model with references. Best at combining multiple characters into a single image. | Multi-character scenes, cinematic style | 1/3 | ~40 |
| Ideogram 3.0 | Highly aesthetic, general-purpose model. Excels at flat and graphic styles. | Flat design, graphic styles | 2/3 | ~55 |
| Flux 1.1 Pro | Advanced yet efficient model from Black Forest Labs. | Efficient advanced generation | 2/3 | ~30 |
| Flux 1.1 Pro Ultra | Black Forest Lab’s highest quality text to image model. | Highest quality BFL output | 2/3 | ~45 |
| Imagen 3 | Google’s previous generation image model. | Legacy Google generation | 1/3 | ~30 |
| Flux.1 Krea | Distilled and open sourced version of Krea 1. Better at text rendering and anatomy. | Text rendering, anatomy, open source | 3/3 | 5 |