Free

A complete media studio on your own GPU.

Create, edit, mix and design in one window, with local models and full MCP support.

All the local models you need, with no setup required.

A sad beagle puppy lying by the front door

Coming soon

See what it does

Windows · NVIDIA GPU, 24 GB recommended (needed for video), 8 GB minimum · Version 1 stays free

TinyPine Studio: an aurora render with its prompt, Auto in Edit and the library strip

Every picture, clip and sound on this site was made in TinyPine

Looking up through a birch grove in autumn
A steam train crossing a snowy viaduct
A lynx in a snowy pine forest
An unfurling fern with dew
An owl gliding between birches
Yoga on a cliff above a fjord
A diver in a kelp forest
a red fox crossing a snowy forest road
Mist over the Swedish archipelago at dawn
A giant waterfall with a rainbow in Iceland
A humpback whale breaching beside a small boat
a climber resting on a granite ledge above the clouds
fresh cinnamon buns on a rustic wooden table
An animated girl and fox under a mushroom
Anime friends on a train at sunset
An old couple holding hands on a jetty
A volcano erupting at night with rivers of lava
A lightning storm over desert dunes
An old man in a hot spring during snowfall
A Sami reindeer herder with the herd at dusk
A Kazakh eagle hunter on horseback in the mountains
northern lights over a pine forest

Free for real, every feature included

Everything on this site is in the free app: no paid tier, no locked features, no trial period, no subscription and no account. For personal and commercial work, and version 1 stays free for good.

Read our promise
One studio

Nine studios in one window, sharing one library

Pick what you came to do. Each area has its own page with what it can do, shown in the app. They share the library, the queue, the undo and the grade, so work moves from one to the next without exporting anything.

How it fits together

Every line is a path the app really has. Pick an area to see what it hands on and what it takes in.

Images

Prompt, Assist, Vary takes and your own LoRAs.

  • To Photo & colourEvery render opens straight in Edit, with versions beside the original.
  • From VideoSnapshot turns any frame into a picture, with its prompt and seed kept.
  • To DesignGenerate designs and elements, or split a picture into layers.
  • To Music & voiceAssist fills in a song, a voice or a sound effect from a short line.
  • From LibraryOne library, one queue and one undo for every workspace.
Open Images
TinyPine Assist

Type a line. Assist writes the rest.

A local language model with 26 styles and a dial from Faithful to Wild. Each example is a real run: the line, what Assist wrote, and what Krea 2 made of it.

The line

“an old fisherman mending his nets at dawn”

Assist · Portrait

an old fisherman mending his nets at dawn. Hands knotted with salt and time, threadbare nets draped low, taut and frayed across his knees. Low-angle close-up, frame tight on labor, off-center to let the tide’s edge breathe. Dawn’s hush over tidal shore, mist curling at water’s edge, light soft and golden from the east, catching grit and cloth. Muted blues, ochres, steel grey, worn by wind and sea. Photographic portrait, natural light, fine grain, shallow depth of field

an old fisherman mending his nets at dawnwaiting for the prompt
Assist describing an attached picture and proposing the next shot
It looks at your work too

An art director that sees the picture

Attach a picture or a video frame and ask what to do next. Assist describes what it sees, proposes the next scene and writes its prompt. The answer here is the one Assist 8B gave about a dancer in a pine forest; the Images page has it doing the same for two fox cubs, and the render that came of it.

  • Pictures and video frames
  • Fills in songs and voices
  • Finds pictures in your library
AI agents · Model Context Protocol

Let an AI agent work in the app

TinyPine Studio has MCP servers for the timeline and for Design. Connect Claude Code, Claude Desktop or any client that speaks MCP, and the agent can do almost everything you can do in the app: make pictures, video, music, voices and sound effects in the app's own queue, edit and grade them, cut a film and export it finished, lay out a page, run the library and train a LoRA. An open timeline or design shows each change within two seconds as an undo step, so Ctrl+Z takes it back, and what you change at the same time is merged with what the agent does. Moving the library stays with you.

claude › tinypine-timeline
  1. create_timelinename: Nordic Nights teaser, width: 1920, height: 1080, fps: 24
  2. add_mediakind: video, name: 0113-video.mp4, start: 0.0, duration: 4.5, with_sound: false
  3. add_mediakind: video, name: 0098-lighthouse.mp4, start: 4.5, duration: 4.0, with_sound: false
  4. add_mediakind: video, name: 0095-surf.mp4, start: 8.5, duration: 4.0, with_sound: false
  5. add_mediakind: video, name: 0099-concert.mp4, start: 12.5, duration: 4.5, with_sound: false
  6. add_mediakind: audio, name: 0101-slate.wav, start: 0, duration: 17.0
  7. add_graphicshape: rect, fit: frame, start: 0, duration: 1.2, fill: #0b1210, fill2: #14201a, fill_type: linear, gradient_angle: 90, below: false, anim_out: fade
  8. add_texttext: Nordic Nights, start: 0.4, duration: 4.0, template: Title on a panel, style: {animIn: words, animOut: fade, font: EB Garamond}, values: {fontSize: 110}
  9. add_texttext: A teaser cut by an AI agent over MCP, start: 9.0, duration: 4.5, template: Name bar, style: {animIn: slide-left, animOut: fade}
  10. add_graphicshape: star, fit: box, start: 13.5, duration: 3.5, width: 0.06, height: 0.1, x: 0.42, y: -0.36, fill: #e0b64a, anim_in: pop, anim_out: fade

121 tools in the timeline server and 19 in the Design server. These are the ones an agent reaches for most, by their real names.

Make picturesTimeline

Render with LoRAs and Enrich, vary a picture, upscale it, fix faces, cut out the subject and edit it with Auto, presets or any Edit slider.

make_imagevary_imageenrich_promptvary_promptassist_chatedit_imageupscale_imageenhance_facesdepth_mapcut_outselect_objectlens_profilephoto_mergeexport_with_edits
Make videoTimeline

A clip from a picture, longer with Extend, slower, steadier, twice the size, or with something tracked in it.

make_videoextend_videoslow_motionstabilize_videoupscale_videotrack_motion
Make soundTimeline

Songs and longer songs, voices and saved voices, sound effects, and the words in any clip.

make_musicextend_musicmake_speechsave_voicemake_sound_effecttranscribeadd_subtitlescheck_scriptcheck_singing
Build the cutTimeline

Timelines, tracks, clips, titles, panels and shapes, placed where the agent says.

create_timelineadd_trackadd_mediaadd_textadd_graphicadd_shadowadd_layersupdate_clipdelete_clipimport_media
Animate and gradeTimeline

Keyframes, text letter by letter, transitions, masks, green screen and effect clips.

set_keyframesset_animatorsapply_text_presetset_transitionset_masksset_chromaadd_effect
Finish and hand offTimeline

See any frame, check the cut, export the finished video, or hand it to Final Cut Pro, Resolve or Premiere.

render_framecheck_before_exportexport_timelineexport_interchangeexport_files
Run the libraryTimeline

Search by what a picture shows, rate, label, sort into bins, follow where a file came from, and restore from the trash.

search_libraryitem_inforateset_titlelabel_itemsmake_binbin_itemslineagewrite_xmpdelete_itemrestore_item
Train a LoRATimeline

Captions, training, a sample per checkpoint and the one to keep.

train_loradatasetauto_captionset_captionlora_samplesprefer_checkpointlist_loras
Queue, models and settingsTimeline

Watch and steer the queue, install a model from the Grove, read and change settings.

list_queuecancel_jobretry_jobpause_queuelist_modelsinstall_modelget_settingsset_settings
Guides for the agentTimeline

The order of work, which model does what, and every option by name.

video_workflowmodel_guidemusic_guidelist_optionslist_presetslist_edit_settings
Lay out a pageDesign

Artboards, text, shapes, pictures, effects, groups, masks and brush strokes, with the artboard exported at full size.

create_designadd_artboardadd_textadd_shapeadd_imageadd_fxset_fxgroup_layersset_maskpaintupdate_layerdelete_layerset_paletterender_previewexport_artboard

Window › Connect an AI agent in the timeline and in Design shows the exact command for Claude Code and the settings for Claude Desktop, with Copy buttons. The servers run on your computer and read only your library.

Native app

Drawn by your graphics card, at any size

TinyPine Studio is a native Windows app written in Rust on GPUI, the GPU-accelerated interface framework made for the Zed editor. Every frame of the interface is drawn with Direct3D 11, and the effects run on the same card. Panels fold for the picture, rows fold step by step instead of wrapping, and dialogs never leave the window. Before every build the app opens each view at four window sizes and four zoom levels and checks that nothing spills out.

Image with Edit at 1600 × 1000, zoom 0

1600 × 1000 app pixels · zoom 100 % · no element outside its box, checked by the app itself

NativeRust and GPUI, no browser inside
Scales960 × 600 to 2400 × 1470, zoom 90 to 120 %
Shows its speedAn fps meter in the header for whatever plays
Keyboard firstEvery shortcut can be changed
Edit · Auto

Five looks from one click

Auto reads the light, the key and the colour of a picture and grades it toward the look you pick: Standard, Cinematic, Dark & moody, Vibrant or Soft & airy. It says why it changed each thing, and you can take back any part. Pick a picture and a look, then drag across it.

Picture
Aim for

Cinematic. Film-like: soft lifted blacks, highlights rolled off, a little darker, quieter colour, a warm lean.

A saxophonist in a smoky 1960s jazz club, developedA saxophonist in a smoky 1960s jazz club, straight from the modelOriginalAuto · Cinematic
Mixer · recording · MIDI · plugins

A DAW under every timeline

A complete audio workstation inside the video editor. A channel strip per track with EQ, compressor, Auto tone, Tame and a limiter, send buses with reverb and delay, and a master that measures loudness against the standard you deliver to. The mix you hear is the mix the export writes. This is the film from the Editing page, playing.

  • MIDI: instruments, a piano roll, .mid in and out, a MIDI input
  • Elastic audio: stretch between markers with pitch kept, quantize to the grid
  • Delay compensation on every path, plugins included
  • Edit single samples: zoom in and draw them with the Pencil
  • VST3 and CLAP plugins as instruments and effects
  • Record and comp takes
  • Automate any setting
  • Export stems
  • Loudness measured against your delivery standard
The TinyPine mixer during playback: channel strips with their effects, two buses, the master and loudness
The EQ window in TinyPine Studio, during playback

EQ

Shape the tone with up to eight bands on a live spectrum. Drag a point to move a band, double-click the graph to add one.

Clips, tracks, buses, master, a sound in the library Latency: None
Low cutLow shelfBellNotchHigh shelfHigh cutFrequency, gain, QSlopes 6 to 72 dB/oct
  • EQ before or after the compressor, one click
  • The curve is drawn on the mixer strip
Workflow

One idea, all the way through

The same evening, the same library: a line becomes a picture, a grade, a clip with sound, a song, a cut and a poster. Every step below is the real app.

Write the ideaTalk it throughRender it with Krea 2Grade itWrite the musicCut it togetherMake the posterShip it
01 / 09

Write the idea

Write it the way you would say it, pick one of 26 styles and how freely Assist may read it.

  • Quality or Fast
  • Seeds, batches and wildcards
A man biting into a lemon
A grandmother doing a flying karate kick
A man chased by an angry goose
A motorbike chase through an alley
A woman foraging mushrooms in an autumn forest
Friends hiking a steep mountain trail, laughing
Friends jumping off a jetty into a lake at sunset
A tearful reunion hug at an airport
Older couple dancing tango on a Buenos Aires street
Bride and groom dancing at a lively Nigerian wedding
Image and video

The same effects on stills, clips, timelines and designs

One recipe, one set of shaders. Grade a still and the clip made from it gets the same look; on the timeline every setting can be keyframed, and in Design an FX layer applies it to the page. On video the light effects read every frame: flares follow the bright sources and grain is new on every frame.

A still with the recipe applied
Still. Exported from Edit.
Clip. The same recipe, exported as MP4. The lens flares, halation and bloom follow the stage lights frame by frame.
Vary

One character, a hundred takes

Vary asks Assist for a new take of the same prompt: another pose, place, light or weather. Who is in the picture and how it is made stay locked, and every take is checked against the original before it renders. Queue as many rounds as you like and let it run over the weekend.

Skateboarder, the originalSkateboarder, take 2Skateboarder, take 3Skateboarder, take 4Skateboarder, take 5Skateboarder, take 6Skateboarder, take 7Skateboarder, take 8Skateboarder, take 9
Original
Video · LTX-2.5

Stills that move, with sound

Each clip below started as one of the pictures on this site. LTX-2.5 animated it and wrote the sound in the same pass. Click a clip to hear it.

Extend

Longer than one take

Extend continues a clip after its end or leads into its start from the clip's own frames, so the motion carries on. The longer clip is a new video beside the source. Slow it down, steady it or track something in it from the same place.

  • After the Out point
  • Before the In point
  • Both at once
  • From the timeline too
Everything in Video
Quality of life

The small things you notice on day two

There are more than sixty. Here are eight; the full list is grouped by what they save you from.

Undo in every workspace, and a trash you can restore from
Edits never touch the original; versions sit beside it
A missing model is a question with sizes, not an error
Measured file size and free disk before an export
The queue reads as a timeline: Queued, Started and Ended
Search by what a picture shows
Shift+S takes a snapshot
Dialogs stay inside the window; the body scrolls, the buttons stay
Trapeze artists flying high under a circus big top
Two women sparring in an old boxing gym
Gospel choir singing joyfully in a sunlit church
Samba dancer in a giant feather headdress at carnival
Skateboarder catching air in a concrete bowl
Runners in a city marathon crossing a bridge at sunrise
Blacksmith hammering a glowing blade with sparks flying
A Day of the Dead parade in Oaxaca
Busy night market food stall in Taipei at night
Chefs cooking in a busy restaurant kitchen during dinner rush
A hiker crossing a rope bridge over a gorge
Built around its models

A short list of models, each one fitted

TinyPine ships a fixed set of models and every part of the app is written for how each of them behaves. There are no versions to match, no nodes to wire and no settings to hunt for.

Assist · Qwen3-VL
Assist speaks Krea's language

Assist is built on Qwen3-VL, the model family Krea 2 reads prompts with, and writes in the order Krea 2 follows best: the subject first, then style, light and setting.

Krea 2
Samplers made for the model

Sampler, schedule and shift are written for Krea 2's flow-matching schedule, with their own defaults for Quality and Fast and one click back to them.

Krea 2 Raw → Turbo
LoRAs trained the right way round

Train on Krea 2 Raw, the undistilled base, and the LoRA is applied on Turbo, the model you render with. You pick the checkpoint from its sample.

LTX-2.5
Video that continues itself

Extend gives LTX-2.5 the clip's own frames as the start of the next part, so the motion carries on. The longer clip is a new video beside the source.

MiniMax-Music3
Songs that can be continued

The app keeps each song's codes, so the model can write more of a song later. Artist names in a style are replaced by their genres.

MOSS-TTS · Chatterbox · Whisper
Voices made sentence by sentence

Speech is made one sentence at a time, so one line can be redone. Whisper, with a Swedish model, turns what is said into subtitles.

SAM 2.1 · Lucida
Selections that find the object

One click selects an object in Design and in masks; on video it follows the object frame by frame. Cut out lifts a subject with soft edges.

Ming design models
Designs as layers

Generate a design, an element or a background on the canvas, or split a flat graphic into separate layers you can edit.

The Grove
One click, the right build

Each model installs from a signed catalog in the precision that suits your card, hash-checked as it downloads, the moment a feature needs it.

Everything on this site runs on your own card

One download for Windows. The Grove installs each model when you first use it, sized to your GPU.

A sad beagle puppy lying by the front door

Coming soon

Built to be trusted

Your work stays yours

Local

Prompts, pictures, video, sound, LoRAs and your library never leave your computer. The app sends a version check, a notice check and an anonymous usage ping you can switch off. Everything it sends

Signed catalog

The Grove catalog is signed with Ed25519 and checked before a model installs. Files are hash-checked as they download.

Clear limits

Prompts for sexual content involving minors, or intimate images of real people without consent, are refused. Adult art is not. The policy

Early access

Start with a line

TinyPine is in active development, so expect frequent updates and a few rough edges. The app tells you when a new version is out. It is free, and version 1 stays free.

A sad beagle puppy lying by the front door

Coming soon

Free · Windows · NVIDIA GPU, 24 GB recommended, 8 GB minimum · video needs 24 GB