How to use CSuite.

CSuite is a native desktop app for macOS, Windows and Linux. Everything runs on your computer — there’s no account to create and no website to log into.

CSuite has a single window with three main areas:

• The menu: the strip of icons down the left side. The CSuite mark at the top opens the Dashboard; below it sit the four modalities (text, image, audio, video), then Chat and Workflows, and finally Models and Account. Hover an icon to see its label.
• The Dashboard: quick actions plus a Files section grouping everything you’ve made — documents, images, audio, video, compositions, workflows and chats. Clicking any card opens it in the workspace that owns it.
• The sidebar: the list of files in your current modality. For example, in the Image workspace it shows every image in your project folder with a small thumbnail and filename.
• The main panel: where you preview, edit and generate content. A model picker sits above the preview and a prompt sits below it, so you can iterate without leaving the workspace.

Each workspace is built around the same idea: pick a model on the right, type a prompt, and the result lands in your project folder and shows up in the sidebar.

The project folder is the heart of CSuite. Everything you generate or edit is saved there as a real file you can open in Finder, File Explorer, or any other app.

• The app starts with a sensible default folder, so there’s nothing to configure up front. A folder pill above the sidebar shows which folder you’re in; click it to switch.
• Each workspace shows the files it can open: the Image workspace lists the images in the folder, the Audio workspace the audio files, and so on.
• Drag files from your desktop straight onto the sidebar list and they’re copied into the folder and opened. You can also drop them into the folder from outside the app and click the refresh button next to the folder pill to pick them up.
• Once a folder holds more than 20 files, a search box appears above the list, with a sort control (newest, oldest, or by name) tucked inside it.
• Switching to a different folder gives you a clean workspace, great for keeping client projects separate.
• Use the ⋯ menu on any file in the sidebar to reveal it in your operating system’s file manager, duplicate it, or delete it.
• Select a file and press Delete or Backspace to remove it. The app shows a confirm dialog first so you can’t lose work by accident.

An AI model is the engine that does the actual generating. CSuite lets you choose from two kinds of providers:

Cloud providers

CSuite connects to two cloud platforms (Runware and Replicate) that run powerful models on their own servers. Through them you reach models from OpenAI, Google, Black Forest Labs, ByteDance, Meta and more, with a single key per platform. CSuite talks to them directly using your own API key. Your prompts never pass through us.

Local providers

Local providers run open-weight models directly on your computer. Once a model is downloaded you can use it offline, with no per-request cost, and nothing leaves your machine. CSuite supports three runtimes:

• Ollama: the easiest way to run text models like Gemma, Llama, Mistral and Qwen. CSuite can either bundle Ollama for you or use an existing Ollama install you already have.
• Hugging Face runtime: runs ONNX-exported open models. CSuite installs the runtime for you on first use.
• GGML: generates images and video locally, built on stable-diffusion.cpp. CSuite downloads the runtime and model weights on first use. It’s the most demanding of the local runtimes, so each model lists the machine it needs.

How well a local model performs depends on your computer’s memory and (on Macs) your Apple silicon chip. For Ollama models, CSuite compares each model’s needs against your machine and blocks the download if it needs more RAM than you have.

Click the Models icon in the menu. It has two tabs.

Defaults

The first tab, and the one you’ll come back to. It lists one row per thing you can make — text, image, video, chat, speech, music, sound effects — with the model each currently uses. Pick a row to change its model and tune that model’s settings.

These are the same picks the model button inside each workspace writes, so changing one here changes it everywhere.

Providers

The second tab is where keys and downloads live — one entry per cloud platform plus Ollama, Hugging Face and GGML. From here you can:

• Add, reveal, replace or remove a cloud provider’s API key.
• Browse the catalog of local models that you can download. Each row shows the model size, what it’s good at, and whether your computer can run it.
• Download a local model: click Download and watch the progress bar. You can keep using the rest of the app while it downloads, and pull several models at once.
• See which models are already installed and how much disk space they’re using.
• Hide models you don’t want cluttering up the picker when you’re generating content.
• Delete models you no longer need to free up disk space.

The text workspace is a rich-text editor for drafting, rewriting and brainstorming with AI assistance. It opens plain text, Markdown and HTML files.

Generating text

Two things sit in the prompt box to make that easier. The wand button rewrites a rough instruction into a sharper one before you send it. Add context attaches up to five reference sources — text files from anywhere on your computer, or a web page by URL — which the model reads as source material rather than as instructions. Instructions you’ve already used on the open document stay as reusable chips above the box, so a follow-up like “now make the intro punchier” knows what came before.

Inline editing tools

Select any piece of text and a small toolbar appears with quick AI actions:

• Shorten: tighten the selection.
• Expand: flesh the selection out with more detail.
• Improve: rewrite the selection for better clarity (also ⌘J / Ctrl+J).
• More: an overflow menu with Fix grammar, Simplify, Summarize and Continue (which writes on from the selection rather than replacing it), plus Translate and Change tone submenus.
• Generate image: in Markdown and HTML documents, turn the selected text into an image that’s inserted on its own line right after it.

Writing tools

The editor is a real word processor, not just a target for generated text:

• Format ribbon: headings, bold, italic, lists and case transforms, from the Format button.
• Slash menu: in Markdown and HTML documents, type / at the start of a line for headings, body text, lists, quotes and code blocks.
• Find: ⌘F / Ctrl+F opens a find bar; Enter and Shift+Enter step through matches.
• Outline: a document map built from your headings, for jumping around a long piece.
• Stats: a footer under the editor with live word, character and rough token counts.
• Export: save the document as Markdown, HTML, plain text or PDF, whatever format the file itself is in.

Inline images and ghost suggestions

You can ask the model to generate an image right inside a document. It appears inline and is saved alongside the text. When you pause typing, the editor can also suggest a line ahead in faint grey; press Tab to accept it all, ⌘→ / Ctrl+→ to take one word, Esc to dismiss, or use the small Accept / Try again / Cancel controls that appear by the caret. The Autocomplete control in the document toolbar sets the pause length — Medium (3s), Short (1.2s) or Off.

Markdown and HTML documents also accept dragged-in media: drop an image, audio or video file from your desktop (or from another CSuite sidebar) into the editor and it’s copied into your project folder and embedded where you dropped it.

Model settings

The model panel shows whatever the model you picked actually supports, and nothing it doesn’t — so one model offers Reasoning depth, another adds Verbosity, another Temperature and Top P. Turn reasoning up for harder questions and the model thinks for longer before replying. Local models have no settings to tune, so the panel is empty for them.

coastline-dawn.png

RunwareFlux 1.1 Pro16:9PNG+ reference

cinematic coastline at golden hour, soft morning mist…Generate

The image workspace is built around a thumbnail grid sidebar and a large preview panel.

Generating an image

The prompt box has a wand button that rewrites a short prompt into a fuller one (subject, style, lighting, composition, mood), and your last few successful prompts sit above it as chips you can click to reuse.

Editing images

A toolbar above the preview holds the editing tools. Each edit overwrites the file in place, or, with Save a new copy ticked, keeps the original and writes a new file alongside it.

• Resize: pick a preset height (240p up to 1080p), a ready-made social size (link preview, Instagram square, Story, X header), or set a custom width and height with an optional aspect-ratio lock. CSuite uses high-quality resampling under the hood.
• Crop: drag a rectangle in the preview, freely or to a fixed aspect ratio.
• Adjust: brightness, contrast and saturation sliders, previewed live before you apply them.
• Rotate: rotate in 90° steps, or flip horizontally or vertically.
• Annotate: draw on the image with a brush, arrows, rectangles or text, with an eraser and per-mark undo.
• Quality: adjust JPEG compression (for JPEG files).
• Format: convert between PNG, JPEG and WebP.

AI touch-ups

The AI menu above the preview runs cloud models on the image you have open: Upscale 2×, Upscale 4×, and Remove background. The result becomes the new selection. These need a cloud provider; the menu is disabled with an explanation if you only have local models.

To re-generate from an existing image, add it as a reference image when you generate (above), or use Chat or a Workflow.

The audio workspace handles voice, music and sound effects. It opens MP3, WAV, FLAC, OGG and AAC files.

What you can generate

• Voice (text-to-speech): type a script, pick a voice model, and get back natural-sounding speech.
• Music: describe a track and let a music model compose it.
• Sound effects: describe a sound (“a rusty door creaking open”) and get a short clip.

Pick which of the three you want from the Type control in the model panel; the model list filters to match, and each type remembers its own model. Music and sound effects get the wand prompt-enhance button and reusable prompt chips — speech doesn’t, because the text you type is spoken verbatim and rewriting it would change the words.

Voices

• Preview a voice: text-to-speech models list their voices with a play button beside the picker, so you can hear one before writing a whole script. The first play of a voice generates a short sample and costs a fraction of a cent; after that it’s cached and free.
• Clone a voice: models that support it get a Voice cloning control — choose an audio sample, type what’s said in it, and the model speaks your script in that voice instead of a preset one.

Editing audio

• Trim: cut the clip down to a specific start and end time.
• Fade: add a fade-in and/or fade-out.
• Volume: turn the clip’s gain up or down.
• Speed: speed the clip up or slow it down.
• Convert: change the file’s format — MP3, WAV, FLAC or OGG.

The built-in player draws the waveform with play, pause and a scrub bar, so you can see and hear the clip before saving. Every tool has a save a new copy toggle, so you can keep the original if you want to.

opening-shot.mp4

Veo 3.1 Fast

start frame

··· generated motion ···

end frame

0s

6s

The video workspace handles MP4, WebM, MOV and AVI files. For stitching multiple clips, audio tracks and titles into a single longer video, see Compositions.

Generating video

Editing video

• Trim: set a start and end time — either by typing them or by parking the playhead and clicking Set. The kept range shows as an amber band on the scrubber.
• Resize: choose a fixed height (240, 360, 480, 720, or 1080) and the width is calculated automatically.
• Crop: crop the frame freely or to a fixed aspect ratio (1:1, 4:3, 16:9, 9:16 and more).
• Speed: 0.5× up to 2×, with the pitch of the audio preserved.
• Audio: pull the soundtrack out into its own MP3 (it shows up in the Audio workspace), or strip it to get a silent clip.
• GIF: export a looping GIF at your choice of frame rate and width. It lands in the Image workspace.
• Format: convert the container between MP4, WebM and MOV.

AI touch-ups

The AI menu above the preview offers Upscale to 1080p or 4K and Remove background (a green-screen cut-out, which works best on talking-head footage). Both run on Replicate, so they need a Replicate key.

Grabbing frames

Two buttons on the transport bar capture the frame under the playhead. The camera saves it as a PNG in your Image workspace. The second one saves it and drops it into the composer’s start frame slot, which is how you extend a clip: grab the last frame, write a prompt for what happens next, generate. Arrow keys step the playhead a second at a time, and Shift with an arrow key steps a single frame.

The Info button shows duration, resolution, frame rate, codecs and bitrate, plus the model and prompt for clips CSuite generated.

espresso-launch16:9chattimelineDownload…

Helio launch

▶

0:03 / 0:09

0s2s4s6s8s

Title

Helio launch

Pre-order

Image

Hero

Generating…

Video

Demo clip

Audio

Voice + music bed

A composition is a separate creation mode for stitching multiple clips, images, audio tracks and titles into a single longer video. Where the regular video workspace produces one short clip at a time, a composition gives you a multi-track timeline and renders the whole thing out to a single MP4 (or WebM, or MOV).

Starting a composition

Drafting with AI

The fastest way to start is to describe what you want. CSuite plans the beats and lays them straight onto the timeline as clips, then generates the images, voice-over, music and SFX in parallel.

Editing the timeline

• Move and resize: drag a clip’s body to move it, drag its right edge to resize. Clips snap to whole-second ticks and to the start and end of every other clip.
• Reorder: drag a clip up or down within its lane to change which one draws on top where they overlap.
• On the canvas: drag a title or image around the preview to reposition it, and pull its corner handles to resize. Guide lines snap it to the centre and the edges.
• Add clips: use the + Text / Image / Video / Audio palette to insert a new clip. The image, video and audio buttons open a picker listing files from your project folder, with thumbnails. You can also drag files in from your desktop and drop them onto a lane or onto the preview.
• Delete: select a clip and press backspace.
• AI follow-up edits: go back to the AI prompt sidebar and ask for a change (“make the music quieter, add a logo at 3 seconds”). CSuite edits only the clips you asked about and leaves the rest of the timeline alone, so each request is a single undo step.

The Layer panel

Double-click any clip — on the timeline or on the canvas — to open its properties in the right-hand panel. What you get depends on what kind of clip it is:

• Titles: the text itself, plus font, size, weight, colour and alignment. The font list is the fonts actually installed on your machine, with a handful of cross-platform picks suggested at the top.
• Images and video: which file, and how it fits the frame. Video also gets a volume slider (drag it to zero to mute) and a Loop toggle — if a clip runs longer than its source, looping repeats the footage and leaving it off slows the footage down to fill the gap.
• Audio: which file, volume, looping, and whether it’s music, voice or a sound effect.
• Timing: start, length, and fade in/out, on every clip.
• Position and Motion: exact x, y, width, height and rotation as percentages, plus a from/to move-scale-rotate animation with an easing curve.
• Generate with AI: on generated image and audio clips, a Regenerate panel pre-filled with the prompt that created them, and a Stop button while an asset is still generating.

Visual coherence

When the AI drafts a composition, it also picks a short visual theme (palette, lighting, mood) and appends it to every image prompt, so the scenes feel like one shoot rather than a collage of unrelated frames. It’s stored in the composition file — nothing for you to manage.

Rendering to a file

When the timeline looks right, click Download to open the export settings.

• Resolution: Native, 480p, 720p, 1080p or 2160p. Aspect ratio is preserved.
• Framerate: Native, 24, 30 or 60 fps.
• Format: MP4 (H.264), WebM (VP9), or MOV (H.264).

The rendered file lands in your project folder as a regular video file alongside any single-clip generations, ready to play, edit further or share.

Promo thread

Llama 3.1 · local

Make a hero image of a coastline at dawn, 16:9

Saved · coastline-dawn.png

Now narrate the scene in a calm voice

ImageAudioVideoask in plain language, chat picks the tool

Chat is a single conversational workspace where you can talk to a model and produce media in the same thread. Click the chat icon in the menu to open it.

Setting up a chat

What the chat can do

• Stream replies from the chat model.
• Generate a new image, sound or video on request: “make me a hero image of a coastline at sunset”.
• Edit the most recent media inline: “crop this to 16:9”, “trim the first two seconds”, “convert to WebM”.
• Take a previous image as a starting point and produce a variant from it.
• Use earlier images as the start and end frames of a new video.
• Run one of your saved workflows by name (“run my podcast intro workflow”) and post the results back into the thread.
• Read documents you attach — PDF, Markdown, HTML or plain text — and answer questions about them.

Working with a thread

• Regenerate the latest reply if it wasn’t what you wanted.
• Edit and resend your last message: everything after it is removed and the turn re-runs from your rewritten version.
• Copy any single message, or the whole conversation as Markdown.
• Stop mid-answer, which genuinely cancels the work in progress rather than just hiding it.

A workflow is a visual diagram that chains models and edits together — a recipe you can run again and again. Great for repeatable jobs like “turn a blog post into a voiced video with subtitles”.

The node types

• Input: your starting point. A typed prompt, an image, an audio clip or a video.
• Model: runs an AI model. Connect inputs to the “prompt”, “instructions” or media handles, and the node generates an output.
• Edit: a deterministic transformation: resize, crop, trim, convert, transcode.
• Combine: merges several files into one. Use it to play tracks together, or to glue clips end-to-end.
• Switch: an A/B selector. Wire several alternatives into it, pick which one flows on, and compare them without rewiring the graph.
• File: saves the final result into your project folder.
• Note: a free-form sticky-note for labelling sections of the canvas. Notes have no inputs or outputs and never block a run.

Building a workflow

CSuite checks the graph before you press Run: a node that can’t possibly succeed (no model picked, nothing connected, no operation chosen) gets a warning badge, and the Run button stays disabled with the reason in its tooltip. Hovering Run also shows a rough cost estimate for one full pass. If a node does fail, its error stays on the node with a Retry button that picks up from there instead of re-running everything.

Building with the AI assistant

You don’t have to wire graphs by hand. The panel beside the canvas takes plain-language instructions and edits the graph for you: “add a step that upscales the image before saving it”, “branch this so I get a 16:9 and a 9:16 version”, or a whole build from scratch. It adds, removes, configures and connects nodes, arranges them, and lands the whole change as a single ⌘Z undo step. The conversation is saved with the workflow.

Reusing prompts and graphs

• Prompt variables: write {{topic}} anywhere in an Input node’s prompt and it grows an input handle by that name. Wire another node into it and its text is substituted in at run time — one prompt template, many inputs.
• Templates: start from a ready-made graph (text to image, image to video with music, a four-style product shot, a podcast intro) instead of a blank canvas.
• Save as template: from a workflow’s menu in the sidebar, keep your own graph as a starting point for the next one.
• Auto arrange: tidy a canvas that’s drifted into a mess, in one click. Undo and redo work throughout (⌘Z / ⌘⇧Z).

Account has five sections: Analytics, License, Settings, Updates and Feedback. Provider keys live under Models, not here.

Analytics

• See total spend, request count, and token usage across the last 7, 30, 90 days, or all time.
• Break it down by provider, model, and modality so you can spot which workflow is eating your budget.
• Local-provider rows show $0.00. They’re free to run, and they’re tracked alongside cloud usage so you can compare apples to apples.

License

• Activate a license: paste your key and click activate. The app validates the key online once, then stores a signed receipt that works offline.
• See your license status: active, in its offline grace window, expired, or moved to another machine, along with when it was last validated. The app re-checks periodically in the background.
• Deactivate a machine: frees up the seat so you can use the license on a different computer.

Settings

App-wide preferences, in one panel:

• Appearance: System, Light or Dark. System follows your operating system.
• Notifications: an OS notification when a generation finishes. It only fires when you’re looking elsewhere, so it never interrupts you watching the thing you’re waiting for.
• Privacy: a single switch for the anonymous daily ping described below, and a link to the full privacy policy.
• Reset: returns the app to a fresh state. It clears your local app configuration and relaunches so you can set it up again. Your generated content stays put; anything in your project folder is left untouched.

Updates

• CSuite keeps itself up to date: it checks for new versions in the background, downloads them quietly, and installs on the next restart. When an update is ready, a Restart to update button appears here, and a small indicator shows in the side menu while one downloads.
• Prefer to stay in control? Untick Automatically download and install updates and use Check for updates whenever it suits you.
• If updating fails repeatedly, the app points you to the website to download the latest version manually.

Feedback

Send bug reports and feature requests straight from the app and track their status over time. Each one can come back with comments as we follow up. Sending feedback requires an activated license.

The anonymous daily ping and its off switch live under Settings, above — see Privacy & data for exactly what it contains.

CSuite is local-first. Here’s exactly where your data goes:

• Your files live in the project folder you picked. They are never uploaded anywhere by CSuite.
• Your prompts go directly to whichever provider you picked for that generation. Local providers stay on your machine; cloud providers receive the prompt through your own API key.
• Your provider keys are stored locally on your computer, encrypted at rest with your operating system’s secure storage (Keychain on macOS, DPAPI on Windows, libsecret on Linux), never on our servers.
• Your license is verified online when you first activate it; after that the app uses a signed receipt that works offline.
• What the app tells our servers: license checks and feedback send only what’s needed to do the job (your license key, app version, platform and a machine identifier); update checks just fetch the release feed; and an optional anonymous daily ping counts active installs — a random id, app version, OS and language, with an off switch in Account → Settings. Nothing else.
• What we never see: your prompts, your generations, your project files, your API keys, or which models you use.

A generation failed

• Check the model picker. Is the right provider selected?
• Open Models, click the provider, and press Save key again to re-validate. Expired or invalid keys are the most common cause.
• Some cloud providers throttle accounts that run out of credit. Check your usage and balance on the provider’s dashboard.

A local model is stuck or slow

• Bigger models need more memory. If your machine has limited RAM, try a smaller model from the catalog.
• Restart the app. Local runtimes free up memory on quit.
• In Models, try deleting and re-pulling the model if a download was interrupted.

A video edit was canceled or didn’t finish

• Quitting the app while a video edit is running cancels the job. Re-run it once the app is back open.
• Very large source files take time. The progress bar shows how far along the edit is.

Still stuck? See Support below for how to get in touch.

CSuite is built and supported by a small team. Every email goes to a real person, and we read every one.

• Email support@csuite.so for bugs, questions, license help, and feature requests.
• Include your operating system, a short description of what you were doing, and (if relevant) the provider and model you had selected. The more we know up front, the faster we can help.
• Lost your license key? Email us from the address you used to purchase and we’ll resend it.
• Have an idea? We genuinely want to hear it. The same inbox is the right place.

CSuite is a one-time purchase, sold with a no-questions-asked refund window. We’d rather you spend money on something you love than feel stuck with something you don’t.

• 7-day refund: if CSuite isn’t the right fit for any reason, email us within 7 days of purchase and we’ll refund your order in full.
• No forms, no surveys: a one-line message is enough. We don’t require a reason, though we’re always grateful when people share one so we can improve the app.
• How it works: refunds are issued to the original payment method via Stripe (typically 5 to 10 business days to land back on your statement). Your license key is deactivated as part of the refund.

To request a refund, email support@csuite.so with the email address you used at checkout.