QueyFramesQueyFrames
gemini aikeyframesaudioblogearly access

node-based media canvas · by QuokkaQuery

An AI video generator you wire together.

QueyFrames puts AI video, image, music, voice and text generation on a single canvas. Each node publishes what it made; pull a wire and the next one consumes it. Everything lands on a composite timeline you can drag, trim and scrub — no exporting, no round-tripping files between four different tools.

Video and stills come from Google’s Gemini models, music and voice from ElevenLabs, and programmatic renders from Remotion. What holds them together is keyframes: shared timing every node can read.

sign up for early access→common questions

Every generator on one canvas

Media nodes lead with what they made — the still, the player, the waveform — with the prompt and parameters underneath. Every generative node talks to a real backend; nothing here is a mock.

  • video

    gemini

    Gemini omni-flash video generation — text-to-video, or image-to-video from a still wired in as the first frame. 3–10 seconds at 720p, 16:9 or 9:16.

  • image

    gemini

    Gemini image generation from a prompt, with model, aspect ratio and size selects. Or upload your own still and use it as a first frame.

  • render

    remotion

    A real server-side Remotion render at 1920×1080, 30fps. Describe the shot and it fills in the composition, the title and every keyframe span for you to edit.

  • music

    elevenlabs

    AI music from a prompt, or section by section from a composition plan — and inpainting to replace a stretch of a track you already like.

  • voice

    elevenlabs

    Text-to-speech with voice and model selects, fed by any text node. Write the script upstream and the voiceover re-renders when it changes.

  • sfx

    elevenlabs

    Sound effects from a description, with trim handles that really cut the clip — what goes downstream is what you trimmed, not the original.

  • keyframes

    on canvas

    Keyframe spans with a description and animatable properties, edited by hand or generated. Each one leaves on its own port, so a single beat can drive a single node.

  • waveform

    in browser

    An analyzer: play a clip back, mark its sections on the wave, and send those timings on as keyframes for the video and render nodes to cut against.

  • timeline

    in browser

    The composite: three tracks, draggable and trimmable clips, a marker lane and a scrubbable ruler with a live preview of the frame under the playhead.

  • llm

    gemini

    Gemini text generation with a model select and a free-text or list output mode. Up to six wired inputs, placed into the prompt by name.

  • palette

    gemini

    Brand colors — describe a palette and it names the swatches, or type them in. Each color has its own output port, so one can feed one node.

  • text

    in browser

    A plain textarea that publishes on every keystroke. The prompt source everything else reads from, so one edit re-times a whole branch.

Keyframes are the thing everything else agrees on

A keyframe here is not one node’s private setting. Mark a beat in a song, a section on a waveform or a moment in a clip, and that timing leaves on a port for anything downstream to read — so motion lands on the beat, cuts line up across layers, and a change upstream re-times everything that depends on it.

  • one clock, every node

    Markers are named milliseconds into the media, never a fraction of a node's own length — so a mark means the same instant wherever it is read.

  • spans, not points

    A keyframe segment carries a start, an end, a description and the properties it animates, so it stands on its own instead of only meaning something relative to the segment before it.

  • generated or hand-cut

    Describe a shot and get a whole plan back — composition, title and keyframe spans — then edit any row. Or drop marks by hand on a waveform or a clip.

How it works

  1. 01 · open a canvas

    A project is a tree of canvases: folders nested as deep as the work needs, and every file is a canvas. Open one and the sidebar lists what is on it.

  2. 02 · add generators

    Drag a node out of the palette docked at the bottom, right-click the canvas to drop one where you clicked, or press shift+A and type its name.

  3. 03 · wire them up

    Pull an output dot into an input socket. Text drives prompts, an image becomes a first frame, keyframes re-time a render. A wire has to carry one kind of data end to end, so a bad connection is refused at the port with the reason.

  4. 04 · composite and scrub

    Every connection into the timeline node becomes a clip, auto-placed in the first free gap. Drag it, trim it, drop marks on the ruler, and scrub the preview.

Or ask for the graph

A Gemini chat sidebar sits beside the canvas and can build it for you: add nodes, rename them, write their prompts, set keyframes, wire them together and lay them out. It runs the same connection rules your pointer does, so it can’t wire anything a hand couldn’t.

Common questions

What is QueyFrames?
QueyFrames is an AI video generator built on a node canvas. Instead of one prompt box, each generator — video, image, music, voice, sound effects, text — is a node you can wire to the others, and everything lands on a composite timeline you can drag, trim and scrub.
Which AI models does it use?
Video, images and text run on Google's Gemini models, including Gemini omni-flash for video generation. Music, text-to-speech and sound effects run on ElevenLabs. Programmatic renders are produced by Remotion on the server.
What makes it different from a single-prompt AI video generator?
Timing is shared. Every generator publishes its keyframes — beats in a song, section bounds on a waveform, marks on a clip — and downstream nodes re-time to them. Change something upstream and everything that depends on it moves with it, instead of being regenerated by hand.
Can it generate video from an image?
Yes. Wire an image node into a video node and its still becomes the first frame for image-to-video generation. You can also send reference images alongside the prompt and name each one by position in the prompt text.
Do I need my own API keys?
No. Generation is metered in credits on your account and every provider call runs server-side, so no keys ever reach the browser. New accounts start with a free credit grant, and credits can be topped up with one-off packs or a Plus subscription.
Is it available now?
QueyFrames is in early access. Create an account to join — the canvas opens once your account is approved.

Keep reading

  • Gemini AI video generator — text to video

    How QueyFrames uses Gemini AI for video and stills — text to video, image to video, and the prompt surface around them.

  • AI keyframe animation for AI video

    What a keyframe means here: a timing every node reads, so motion, cuts and audio land on the same beat.

  • AI music, voice and sound effects generator

    The audio half — AI music, AI voice generation and sound effects, all publishing timings the video nodes read.

  • AI video generation guides and workflows

    Guides and workflow notes — prompting for pacing, image to video, and keeping audio and picture on the same beat.

A blank canvas and a palette of generators.

QueyFrames is in early access. Create an account to join the list.

sign up for early access→

QueyFrames — AI video generator by QuokkaQuery

overviewgemini aikeyframesaudioblogtermsprivacysign up