> For the complete documentation index, see [llms.txt](https://help.intangible.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://help.intangible.ai/learn-more/faq/known-limitations.md). # Known limitations What Intangible doesn't do. Useful for setting expectations on a brief, scoping a pitch, or knowing when to use external tools. Each entry names something Intangible doesn't do and gives the reason or the workaround, so you can scope a brief or a pitch around it before you promise the work. ## Canvas and Chat ### The status can say done before the work is The status line in the chat panel can read "All done! How does it look?" while steps are still unfinished. Treat the step list and the canvas as the record of what was built. **Workaround**: check the step list and look for the cards each step makes. If steps are left, ask the agents to continue. See [Troubleshooting Intangible agents](/learn-more/faq/troubleshooting-the-agent.md#the-status-says-done-but-the-canvas-is-missing-pieces). ### Canvas cards can't be attached to a chat message Selecting a card on the canvas does not pass it to the chat panel, and the chat panel **+** attaches images only. **Workaround**: name the card in your message the way the canvas names it, for example "shot 02's keyframe". See [Chat with Intangible](/chat-agents-intangible/chat-with-intangible.md#point-at-what-you-mean). ### Cards can't be multi-selected You select one card at a time on the canvas. **Workaround**: to act on several shots at once, name them all in one message to the agents ("Render shots 02, 03 and 05 as video"), or use [Batch visualize](/canvas/batch-visualize.md) to render every shot in a scene. ### Videos the agents start skip the cost notice The Visualize panel shows a notice with the credit count above **Generate** for a large video. A video the agents start goes straight to the job, so you never see that notice. The agents quote the credits in the chat first and start the video only when you agree. **Workaround**: check the credits the agents quote before you agree, or render the video yourself from Visualize. See [Keyframes to video](/canvas/keyframes-to-video.md#credits). ### No running cost total or time estimate The agents report the cost of a step when they start it, for example a set of reference sheets. There is no running cost total for the session and no time estimate for a run. The rotating line under the latest reply shows the run is alive, not how far along it is. **Workaround**: watch the step list for progress, and check the credits button at the top right for your balance. See [Credits and tokens](/learn-more/concepts/credits-and-tokens.md). A job that fails is refunded automatically. Results that complete but miss the mark are charged. ### Video settings from the agents need checking The agents offer video options with their own model and length, and the clip length may not match the shot. A video animated from a render invents its motion; it does not follow the scene's animation unless you pick the option that does. A shot or source video shorter than the model's minimum is refused before any credits are spent, with a message that names the minimum. **Workaround**: pick the option that follows the scene's motion, or say what the video should start from ("from its keyframe and scene"). Then open the shot in Visualize, where the agents write their settings, and adjust them before you render again. See [Generate video](/visualize/generate-video.md). ### Files can't be dropped onto the canvas Drag-and-drop from your computer onto the canvas does not upload. **Workaround**: click **Upload media from local** in the bottom toolbar, or upload into a media group. To give images to the agents, attach them with **+** in the chat panel, or drop them on the message field in the chat panel. See [Create on the canvas](/canvas/create-on-the-canvas.md#upload). ### No frame-selected shortcut on the canvas Pressing **F** frames a selected object in Build, but not a card on the canvas. **Workaround**: zoom with the scroll wheel, and pan by dragging the empty canvas, or by dragging with the right or middle mouse button. See [Groups and cards](/canvas/groups-and-cards.md#move-around-the-canvas). ### The full-size view shows one card at a time Double-clicking an image or video opens it full size, with no previous or next control. **Workaround**: press **Esc** to close it and double-click the next card. ### New renders don't get their own view Visualizations lists the newest renders first, but the canvas does not move to them as they land, and there is no view of just the latest batch. **Workaround**: look in **Visualizations**, or zoom out to see the whole project. ### The chat panel is on the canvas and in Build only Compose and Visualize have no chat panel. **Workaround**: ask from the canvas or Build; the agents can still move cameras, set keyframes and render. See [Chat with Intangible](/chat-agents-intangible/chat-with-intangible.md). ### Undo does not cover canvas card moves Cmd/Ctrl+Z undoes scene edits, yours and the agents', but not many canvas card actions, such as placing uploads or adding cards to groups. A page reload clears the undo history. **Workaround**: delete or move the card by hand. ### No live multiplayer Two people can edit one project, but not side by side in real time. **Workaround**: agree who edits which assets, and avoid editing the same ones at once. See [Sharing and collaboration](/learn-more/reference/sharing-and-collaboration.md). ### No crop tool in Edit Edit marks up an image for the model to change; it does not crop. **Workaround**: download the image and crop it in the image editor you already use, such as Photoshop, Affinity Photo or your computer's built-in photo app. ### No video editor Edit changes what happens in a clip; it does not trim, join or reorder clips, and there is no editing timeline for titles, transitions or sound. **Workaround**: join shots by exporting the scene or sequence, which cuts them together in order (see [Export Video](/publish/export-video.md)). For everything else, download the clips and cut them in the video editor you already use, such as Premiere Pro, DaVinci Resolve, Final Cut Pro or CapCut. ## Output and formats ### FBX export isn't supported Scenes export as GLTF or USD, not FBX. **Workaround**: use USD or GLTF; Maya, 3ds Max, Blender and Houdini import one or both. See [Export formats](/publish/export-formats.md). ### Audio mixing across shots isn't supported Each shot's video output carries its own audio if you generated with audio on. The Sequence tab plays only the active shot's audio – no blending or crossfading across shots. Video export keeps per-shot audio rather than producing a continuous mix. **Workaround**: edit audio in your editing tool downstream. Premiere, Resolve, Pro Tools. ### Image output is PNG; scene references are JPG Rendered stills export as PNG. Reference images attached to a 3D object in the scene store as JPG. The distinction matters only if you're round-tripping the assets externally. ### Higher-bit-depth and HDR formats aren't exposed PNG and JPEG are the export options for stills. TIFF, EXR, and other production-pipeline formats aren't on the list. **Workaround**: PNG covers most cases. For HDR / EXR specifically, no equivalent. ## Generation ### Voice input needs a supported browser The microphone on Home and in the chat panel shows only in browsers that support speech recognition. Where it is missing, the browser can't take dictation. **Workaround**: type the prompt, or open Intangible in a browser that supports voice input, such as Chrome. ### Custom diffusion parameters (seed, steps, edge retention) aren't exposed in the UI The current UI surfaces only Scene / Style / Lighting tabs. Advanced visualizer parameters aren't available to users. **Workaround**: pick the right model, pick the right preset, attach an image reference. Most of what advanced users want to do via these parameters can be approximated via the visible controls. ## Workflow ### Cross-scene shot edits aren't supported A single shot's camera can't span two scenes. To cut between scenes, build separate shots in each scene and assemble in the [Sequence Timeline](/compose/timeline.md). **Workaround**: assemble in the sequence timeline. The sequence cuts cleanly across scenes. ### Cross-team sharing isn't supported A project, custom style, or saved set in Team A's workspace isn't visible to Team B. There's no cross-team sharing flow. **Workaround**: re-create in the other team. Or share at the project level via email ([Sharing and collaboration](/learn-more/reference/sharing-and-collaboration.md)). ### iPad support isn't fully tested Intangible Studio runs on iPad, but not every function has been verified for touch. Viewport gestures and flows that depend on precise pointer input may behave unevenly. **Workaround**: use a desktop or laptop browser for production work. iPad is fine for review and light edits. ## Performance ### Big model imports are capped and slow A model import is rejected over 512 MB or 2 million vertices, and one close to those limits imports slowly. **Workaround**: decimate externally before importing. A clean retopology of the source mesh imports faster and renders better. ### Very large scenes get sluggish Past several thousand objects, the viewport's frame rate drops on most hardware. **Workaround**: use [populators](/build/populators.md) for dense distributions instead of individual placement. A single populator with a thousand procedural instances performs much better than a thousand individual objects. ## Generation models ### Some video models cap aspect ratios Vertical 9:16 and square 1:1 aspect ratios may not be available on every video model. The model dropdown filters as you pick a gate. **Workaround**: pick a model that supports the gate, or render at a supported gate and crop externally. ### Provider models change The visualizer is a model aggregator; provider lineups evolve. Models you used last quarter may be replaced or retired. **Workaround**: the visualizer's dropdown is authoritative. Pick the closest current successor when a model disappears, and re-render to confirm the look survives. ## What to do when you hit a limit Intangible exports standard files (PNG and JPEG stills, MP4 video, GLTF and USD scenes), so finishing work it doesn't do, such as cropping, cutting, titles, colour or sound, goes into the tools you already use. If a limit is blocking work, send a note via [Report a bug or get help](/get-started/help-and-feedback.md). The team tracks limitations and prioritizes them based on how often they surface in real work. {% hint style="info" %} **Still stuck, or think it's a bug?** Report it from inside Studio: click the help icon (?) in the top-right toolbar and pick **Report an Issue**. Describe it, click **Send report** and share the Studio tab for a screenshot. See [Report a bug or get help](/get-started/help-and-feedback.md). {% endhint %} ## Related * [Common questions](/learn-more/faq/common-questions.md) * [Troubleshooting rendering](/learn-more/faq/troubleshooting-rendering.md) * [Troubleshooting import](/learn-more/faq/troubleshooting-import.md) * [Report a bug or get help](/get-started/help-and-feedback.md) --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://help.intangible.ai/learn-more/faq/known-limitations.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.