Tool index19 tools21 modelsUpdated November 2026

Pick a tool. Give it a job. Get it done. AI tools for images, video, audio and documents

Generate images and video, turn recordings into text, give text a voice, question a PDF, and merge, sign or compress one. One credit balance, no subscription.

01

Tools

Every tool runs on the same credits. Open one and start; nothing to install.

How credits work →
02

Image models

From a quick draft to a finished piece, and editing by instruction. Point at a name to see what it makes.

All 15 image models →
03

Questions

Not answered here? Ask us →

What tools are on onaiagents?

An AI image generator and video generator, audio and video transcription, text to speech with 23 voices, chat with PDF, a , an image editor that works by instruction, image to video, Visual Deconstruction (image to prompt), AI Friends, and a model directory with prices. More tools are on the way, and all of them run on the same credits.

Can I try the tools without an account?

Yes. Audio to text, text to speech, chat with PDF and the give you 15 free runs between them without signing in. After that, credits keep them going.

Is there a subscription?

No. You buy credits once and spend them on whichever tool you like. Credits never expire, and every new account starts with free credits. See pricing.

Which AI models can I use on onaiagents?

Grok Imagine 2.0, Flux Krea, Qwen Image 2512, Z-Image, HunyuanImage 2.1, HiDream I1, Krea 2 and Bria FIBO for images; Qwen Image Edit, Flux Kontext and Flux 2 Klein for editing; Grok Imagine Video, Wan 2.2, LTX-2 and HunyuanVideo 1.5 for video. The model directory lists every one with its price.

Can I edit an image I already have?

Yes. Open Qwen Image Edit or Flux Kontext, attach the picture, and describe the change. Everything you did not mention is kept.

Can I turn an image into a video?

Yes. Pick an image-to-video model such as LTX-2, Wan 2.2 or Grok Imagine Video in the video generator, attach a still, and describe the motion.

What is Visual Deconstruction?

Upload any image and Deconstruct writes the prompt that would recreate it: subject, lens, lighting, palette and composition, measured rather than guessed.