AI models you can use in FARZI Studio
FARZI Studio puts 10 AI video models and 11 AI image models - from Google, ByteDance, Kuaishou, Black Forest Labs, OpenAI, xAI, Alibaba and others - in one browser workspace, next to Avatar Studio and the AI Video Editor. This page lists each one with what it supports, so you can pick the right model before you start.
Start in FARZI Studio How pricing works
Updated from the live catalogue on 2026-10-29. Options can vary by mode; the app shows exactly what each model offers for your settings.
AI video models
Text to video, image to video and video from reference images. Length and resolution depend on the model.
| Model | Developer | Modes | Resolutions | Length | Sound |
|---|---|---|---|---|---|
| Gemini Omni Flash | Text to video, Image to video, References to video | 720p, 1080p | 6-10 s | Yes | |
| Grok Imagine Video 1.5 | xAI | Text to video, References to video | 480p, 720p | 4-15 s | Yes |
| Seedance 2.0 Fast | ByteDance (Seed) | Text to video, References to video | 720p | 5-15 s | Yes |
| Seedance 2.5 | ByteDance (Seed) | Text to video, References to video | 480p, 720p, 1080p | 4-30 s | Yes |
| Seedance 2.5 Fixed 30s | ByteDance (Seed) | Text to video, References to video | 720p | 30 s | Yes |
| Veo 3.1 Fast | Text to video, Image to video, References to video | 720p, 1080p | 8 s | Yes | |
| Wan 2.2 | Alibaba (Tongyi Wan) | Text to video, Image to video | 480p, 720p, 1080p | 5 s | No |
| Wan 2.2 Turbo | Alibaba (Tongyi Wan) | Text to video, Image to video | 480p, 720p, 1080p | 5 s | No |
| Wan 3.0 | Alibaba (Tongyi Wan) | Text to video, Image to video, References to video | 480p, 720p, 1080p | 2-30 s | No |
| Wan 3.0 Prime | Alibaba (Tongyi Wan) | Text to video, Image to video, References to video | 480p, 720p, 1080p | 2-30 s | No |
AI image models
Create images from a prompt, or edit and combine pictures with reference images.
| Model | Developer | Modes | Resolutions |
|---|---|---|---|
| Gemini 3 Pro Image | Text to image, Image editing | 1k, 2k, 4k | |
| Gemini 3.1 Flash Image | Text to image, Image editing | 1k, 2k, 4k | |
| GPT Image 2 | OpenAI | Text to image, Image editing | 1k, 2k, 4k |
| GPT Image 2.5 Flare | OpenAI | Text to image, Image editing | 1k, 2k, 4k |
| GPT Image 2.5 Sunburst | OpenAI | Text to image, Image editing | 1k, 2k, 4k |
| Grok Imagine Image | xAI | Text to image, Image editing | 1k, 2k |
| Grok Imagine Image Quality | xAI | Text to image, Image editing | 1k, 2k |
| Nano Banana 2.1 | Text to image, Image editing | 1k, 2k, 4k | |
| Qwen Image 2512 | Alibaba (Qwen) | Text to image | shown in the app |
| Qwen Image Edit 2511 | Alibaba (Qwen) | Image editing | shown in the app |
| Z-Image Turbo | Alibaba (Tongyi Lab) | Text to image, Image editing | shown in the app |
Avatar and head-swap tools
In Avatar Studio: a portrait that talks or sings your audio, or a head replaced in a video.
| Tool | Developer | Input | Resolutions | Input length |
|---|---|---|---|---|
| InfiniteTalk | MeiGen-AI | Portrait + audio / video | 480p, 720p | 3-600 s |
| LongCat Avatar 1.5 | Meituan (LongCat) | Portrait + audio / video | 480p, 720p | 3-64 s |
| Video Head Swap | - | Portrait + audio / video | 480p, 720p | 3-120 s |
Pricing
FARZI Studio is pay as you go in US dollars: you add balance to your wallet and each generation is charged from it - there is no subscription to buy.
Before anything is generated the app shows the exact price for your settings. That is the amount taken from your balance, and if a generation fails it is returned in full.
Frequently asked questions
Which AI models can I use in FARZI Studio?
10 video models and 11 image models, plus the Avatar Studio tools listed above. They all run in one workspace, so you can compare them on the same prompt.
How do I choose a model?
Start from what you need: text to video or image to video, the length, the resolution and whether you want sound. The tables show what each model supports; the app only offers settings the model really supports.
How much does each model cost?
The price depends on the model and your settings (length, resolution, sound). The exact price appears in the app before you generate, and that is the amount charged; a failed generation is returned in full.
Are the tools on this page always available?
This page is updated from the live catalogue. A model can be paused while a provider is unavailable; the app always shows what you can use right now.
Who made these models?
Each model is developed by the company named in the table. FARZI Studio gives you access to them in one place; the names are trademarks of their owners.
Related tools
Try these models in FARZI Studio
Open FARZI Studio in your browser - nothing to install.
