Skip to content
ReelMarkr Docs

AI overview

ReelMarkr’s AI is local-first. Image generation, the assistant, dictation, and transcription run on your machine with no account and no per-use fees. You can optionally connect cloud models with your own API keys: the app calls the provider directly, and the provider bills you.

  • Image generation: a node-based graph (ControlNet, LoRA, img2img, inpaint, outpaint, upscale, face-fix, background removal, compositing) on your GPU. See Image generation.
  • Cloud AI (optional, bring your own keys): cloud image models and video generation nodes (Kling, Veo, Seedance, and more), with cost estimates and a spend ledger on every run. See Cloud AI.
  • AI assistant: chat with a local LLM via Ollama or LM Studio, or a cloud provider, that builds on and organizes your canvas. See AI assistant.
  • Dictation: local voice-to-text into notes, documents, and the chat via Whisper on your CPU. See Dictation.
  • Voice memos: record audio onto the board and transcribe it locally. See Voice memos.