AI models on Vofy
Browse image and video models, compare capabilities, and open any model to create with it.
Image models
Compare image models for generation, editing, reference workflows, product shots, and design work.

GPT Image 2.5
GPT Image 2.5 is OpenAI's image generation and editing model, built for high prompt fidelity and precise control. On Vofy it renders from 1K up to 4K, supports inpainting on an existing image, and accepts exact output dimensions instead of a fixed set of aspect ratios.
Seedream 5.0 Pro
Seedream 5.0 Pro is ByteDance Seed Team's multimodal image creation model for professional design work. It improves image-text alignment, structural coherence, text rendering, visual aesthetics, dense information layouts, precision editing, realistic textures, and multilingual generation.
Nano Banana 2 Lite
Nano Banana 2 Lite is the streamlined 1K edition of Nano Banana 2 for fast image generation and editing. It keeps the Gemini image workflow focused on responsive drafts, references, and everyday creative edits.
Grok Imagine Image Quality
Grok Imagine Image Quality is xAI's recommended higher-quality image model replacing the retiring Pro tier. On Vofy, it supports prompt-based creation, image edits, broad style transfer, and multi-turn refinement at up to 2K with up to 10 outputs per run.

GPT Image 2
GPT Image 2 is OpenAI's state-of-the-art image generation model for fast, high-quality image generation and editing. OpenAI positions it as a major step forward in instruction following, dense text rendering, multilingual layouts, stylistic fidelity, flexible sizing, and stronger world knowledge.
Nano Banana 2
Nano Banana 2 combines Pro-level image quality with Gemini Flash speed — advanced world knowledge, subject consistency across 5 characters, precise text rendering and translation, and 4K output from 512px, all powered by real-time web search.
Video models
Compare video models for text-to-video, image-to-video, motion control, short clips, and cinematic drafts.
MiniMax H3
MiniMax H3 is MiniMax's high-resolution video model for text-to-video, frame-guided generation and multimodal reference work. On Vofy it renders at 768p or 2K and supports clips from 4 to 15 seconds, with reference images up to 5760 pixels on the long edge.
Wan 3.0
Wan 3.0 is Alibaba's unified AI video model, bringing text-to-video, first and last frame guidance, multimodal reference, video editing and video extension into one generator. On Vofy it renders at 480p, 720p or 1080p and supports single-pass clips from 2 to 30 seconds.
Wan 2.6 Flash
Wan 2.6 Flash is a fast AI reference-to-video generator that creates consistent, multi-shot videos from image or video references, with optional audio.
Gemini Omni Flash
Gemini Omni Flash brings Gemini's native multimodal intelligence to AI video creation and editing. Create 10-second videos from text, photos, references, or source footage, then refine the result with natural-language direction, native audio, and step-by-step creative control.
Seedance 2.0 Mini
Seedance 2.0 Mini is a streamlined Seedance 2.0 video model for short-form generation, reference-guided motion, video edits, and clip extension. It supports text-to-video, image-to-video, first-and-last-frame generation, multimodal references, and audio-visual sync workflows.
Kling 2.6
Kling 2.6 is a balanced Kling video model on Vofy for short clips, motion-controlled video, interpolation, and optional audio workflows. The current Vofy setup supports text-to-video, image-to-video, interpolation, and motion control at 720p or 1080p.
How to choose an AI model
Pick by output type, input support, and the quality-speed tradeoff your project needs.
Start with the output
Choose image models for still visuals and video models for motion, animation, or clips.
Check input support
Look for text prompts, reference images, image editing, image-to-video, or video edits.
Balance quality and speed
Use lighter models for drafts and higher-quality models for final creative work.
FAQ
Which AI image model is best for product photos?+
GPT Image and Nano Banana Pro are strong defaults for photorealism, clean composition, and text-heavy layouts. Seedream is useful when you need fast lifestyle or product variations.
Which AI video model should I use for short films?+
Sora Pro is a strong choice for narrative scenes with audio. Kling is useful for multi-shot sequences, while Veo and Seedance are practical for polished image-to-video motion.
Can I switch models in the middle of a project?+
Yes. Vofy keeps your prompts, references, and project context together so you can compare models without rebuilding the brief.
Do all models support reference images?+
Support varies by model. Image models commonly support reference-based edits, and several video models support image-to-video workflows. Each model detail page lists the relevant inputs.
How are credits charged across models?+
Credits depend on model tier, resolution, duration, and generation mode. The pricing page explains the credit cost for common image and video workflows.
How quickly do new models arrive on Vofy?+
Vofy is designed to add new frontier model releases quickly, so teams can try new OpenAI, Google, xAI, ByteDance, and Kuaishou models from the same workspace.
Ready to create with these models?
Create with Vofy using the image or video model that fits your next project.

