Alle 240+ KI-Modelle
Jedes Bild-, Video-, Audio-, Dokument- und 3D-Modell, das bei faktry verfügbar ist - an einem Ort.
Elevenlabs Speech To Text Scribe V2
ElevenLabs Scribe v2 - Advanced speech-to-text with speaker diarization
Elevenlabs TTS Eleven V3
ElevenLabs TTS v3 - High quality voices
Qwen 3 TTS Clone Voice 1.7B
Qwen 3 TTS Voice Cloning 1.7B - Clone a voice from a short clean speech sample (up to 5 minutes). Saves a voice file to your /voices folder to pick in Generate Speech with Qwen 3 TTS 1.7B.
Minimax Music 3
MiniMax Music 3 - Generate a song from lyrics and a music description
Google Gemini Omni Flash
Gemini Omni Flash v1.1 model for text-to-video generation with in-prompt audio control, up to 4K resolution
Google Gemini Omni Flash
Gemini Omni Flash v1.1 model for animating a single reference image into video, with optional end-frame interpolation, up to 4K resolution
Google Gemini Omni Flash Edit
Gemini Omni Flash v1.1 model for instruction-based video editing (e.g. "Make this video anime. Keep everything else the same."), up to 4K resolution
Topaz Generative
Topaz Starlight diffusion-based video upscaling that regenerates lost detail — Precise/HQ/Mini/Sharp for restoration, Fast 2 for cheaper turnaround
Kling Video Ai Avatar V2 Standard
Standard AI Avatar V2 — animate a portrait image with audio using Kling
Lightricks LTX 2.5 Audio To Video Pro
LTX 2.5 Pro model for audio-driven video generation — animates a starting image to match the audio, or generates from prompt alone
Kling Video V3 Standard Motion Control
Kling Video v3 Standard model for motion control in video generation
Nano Banana 2
Nano Banana 2 (Gemini 3.1 Flash) model for fast and efficient text-to-image generation
Nano Banana 2 Edit
Nano Banana 2 model for fast and efficient image editing
GPT 5.6 Terra
OpenAI GPT-5.6 Terra model for fast and efficient image analysis
Claude Sonnet 5 5
Anthropic's Claude Sonnet 5.5 model for reverse-engineering image prompts
Pixelcut Background Removal
Pixelcut's fast, ultra high-quality background remover, ideal for e-commerce and image editing workflows
Topaz Generative
Middle tier - Topaz Generative rebuilds missing detail, with Wonder 3/3.5 realism, Redefine prompt-guided detail, and Recovery for extreme low-resolution sources. Wonder 3 and 3.5 bill at half the rate of the other variants.
Seedream V5 Flash Layerize
ByteDance's Seedream 5.0 Flash model for fast splitting of an image into a base image and up to 16 separated, independently editable layers
Claude Sonnet 5 5
Anthropic's Claude Sonnet 5.5 model for balanced performance and cost in text generation
Meshy V6
Meshy 6 – realistic production-ready 3D from text
Meshy V7
Meshy 7 – latest-generation 3D from a single image, with optional rigging and animation