Generate Video

The Generate Video node turns a prompt, an image or a set of references into a video clip. It is the main video node in Nodaro: the same node makes text-to-video, image-to-video, first-and-last-frame and reference-driven clips, and it chooses the mode from what you connect. You choose one of the video models, and the result can feed any other video node, for example Combine Videos.

When to use it

• You need a short clip from a description: B-roll, an establishing shot, a product spin, a social post.
• You have a still image and want it to move. Connect it to Start Frame, for example a result of Generate Image.
• You want a clip that starts on one picture and ends on another. Connect both Start Frame and End Frame.
• You want the same character, product or place in many clips. Connect asset nodes or reference images.
• You want to change an existing clip by instruction. Seedance 2.5, Gemini Omni and Gemini Omni Flash 1.2 can edit a connected video.

For a clip longer than one model run allows, use Generate Video Pro. To continue a clip you already have, use Extend Video.

Quick start

How the node chooses a mode

There is no mode switch. When the node runs, it looks at what is connected and picks the mode:

Every model in the list can be chosen in every mode. A few things work differently on some models:

• Some models need an image. Kling 2.1 Master, Kling 3 Omni, Hailuo 2.3, Hailuo 2.3 Pro, Bytedance Pro Fast, HappyHorse 1.1 Ref2V, Vidu Q4 and Grok Imagine Video 1.5 have no text-to-video mode. Run stays disabled until an image is connected, and the button's tooltip names the model. On HappyHorse 1.1 Ref2V and Vidu Q4, reference images alone are enough: the model runs from them without a start frame.
• References without a start frame. Some models run with references connected and no start frame. They are the Seedance 2 family, MiniMax H3, Wan 3.0, Gemini Omni, Gemini Omni Flash 1.2, VEO 3.1 Fast and Lite, Grok Imagine, HappyHorse 1.1 Ref2V, Vidu Q4, Kling 4.0 and Kling 4.0 Flash. Kling 4.0 Keyframes also runs its in-between frames without a start frame. This works in the editor, in a workflow or app run, and through the API. Grok Imagine Video 1.5 is different: its in-between frames need a start frame.
• An end frame alone becomes the start image. If you connect End Frame without Start Frame, the node sends that image as the image to animate.
• One entry, two versions. Grok Imagine, Wan 2.6, Wan 2.7 and HappyHorse 1.1 each appear once in the model list. The node uses the text-to-video or the image-to-video version of the model, depending on whether an image is connected.

Inputs

Generate Video has eleven inputs on its left edge. Each input accepts only the kinds of nodes listed, and the editor rejects a connection that does not fit. An input that the chosen model cannot use is shown disabled. Click an input to see what is connected to it, jump to a connected node, disconnect it, or add a new compatible node.

The output, Video, is the URL of the generated clip. It can feed any number of nodes at once.

Audio Refs is the node's only audio input. It goes into the model and changes what it generates. To lay a soundtrack over a finished clip, connect the clip and the audio to a Merge Video & Audio node.

Workflows saved with the old Audio input. Generate Video used to have a separate Audio input that laid a track over the result. A workflow saved with a connection to it opens with a Merge Video & Audio node added after the clip. The track plays at full volume, and the clip's own sound is kept at 30% only when the model made sound. On LTX 2.3 Pro, the connection moves to Audio Refs instead. On LTX 2.3 Fast, which never used it, the connection is removed.

Typed prompts and connected prompts

A prompt connected to Prompt is added after the prompt you type in the node. If the typed prompt names the connected node as a variable, such as {Subject}, the connected text appears only where the variable is. To stop adding connected prompts, turn off Inject Prompt under Prompt Injection in the settings panel.

A connection to Prompt turns grey and dashed only when the node does not use the connected text. Point at a grey connection to read the reason:

• Inject Prompt is off, and the typed prompt does not name the connected node as a variable. The connection says "Not used: Inject Prompt is off".
• Several text nodes are connected to Prompt. The node uses the one connected last, and the earlier connections turn grey. They say "Not used: a text node connected after this one takes its place". A text node that the typed prompt names as a variable is always used.

Settings input

The Settings input takes the Aspect Ratio, Duration, Provider and Motion nodes. When the node runs, each connected Aspect Ratio, Duration or Provider node replaces one of the node's own settings:

A Motion node replaces no setting. It adds a clause about the amount of motion to the end of the prompt, with the wording of the connected pickers. For example, Dynamic adds dynamic, energetic motion with fast, pronounced movement. The clause follows the Inject Look switch under Prompt Injection, like the Look pickers. A connected Motion node does not turn off the node's own Motion hint (injected into prompt) setting, and the two are not merged, so use one of them. See Motion.

• Several nodes of one kind. When two nodes of the same kind are connected, such as two Duration nodes, the one connected last is used.
• A model the node cannot run. A Provider node set to an image model is refused. The model's chip on the node turns red, and Run is disabled with the reason, for example "This node can't run Nano Banana. Choose one of its models in the Provider node." A run that reaches the node, for example from a workflow run, stops there before anything is charged, and the error names the Provider node.
• What the node shows. Chips at the top left of the node list the values it runs with, for example 16:9, 8s, Seedance 2 and Dynamic. In the strip under the node and in the settings panel, a connected Aspect Ratio, Duration or Provider shows its source node and cannot be changed there.
• The price. The Run button, the workflow total and the credits held when the run starts all use the connected model and the fitted length. A Motion node does not change the price. While a setting is connected, a note under Run says "Price may change with connected settings".
• Adaptive and Auto. A connected ratio replaces Adaptive and Auto (from image), and a connected length replaces Auto.

Settings

These settings appear for most models. The settings panel shows only the controls the chosen model supports.

Provider, Duration and Aspect Ratio nodes connected to the Settings input replace Provider, Duration (seconds) and Aspect Ratio when the node runs. A connected Motion node adds a clause to the prompt instead. See Settings input.

Settings for specific models

Kling 3.0 settings

When you choose Kling 3.0, the settings panel changes to three tabs: Scene, Shots and Elements.

• Scene holds the model, the Mode (Pro (1080p), Standard (720p) or 4K (Ultra HD)), the Aspect Ratio, Sound Effects for lip-synced dialogue and sound, and the duration from 3 to 15 seconds.
• Shots has Multi-Shot Mode, which splits the clip into 2 to 6 shots, each with its own prompt and length, up to 15 seconds in total. Sound is required in multi-shot mode, and an end frame is not supported.
• Elements lets you define named characters or objects with a description and 2 to 4 reference images, then mention them in the prompt as @name.

Which model to choose

Generate Video can run every model below. Click a model for its settings, credit prices and prompt tips.

• Seedance 2.0 Fast — the default. Low cost, sound on, references supported. Use it to iterate.
• VEO 3.1 Fast — a good balance of price and quality, with native audio and the same price for 4, 6 or 8 seconds.
• VEO 3.1 Quality and Kling 3.0 — premium cinematic shots. Kling 3.0 adds multi-shot clips and lengths from 3 to 15 seconds.
• Seedance 2 and Seedance 2.5 — the widest reference support: images, videos and audio together. Seedance 2.5 makes up to 30 seconds in one run and can edit a clip.
• Wan 3.0 — 2 to 30 seconds at any whole second, with references and ambient audio.
• Vidu Q4 — a start frame, or up to 15 reference images and 3 voice or sound clips, with native audio, from 540p to 4K.
• Kling 4.0 — up to 30 seconds with native audio, from reference images, videos, characters and voice clips, or from text alone. Kling 4.0 Keyframes moves through a start frame, up to 8 in-between frames and an end frame.
• Kling 4.0 Flash — the faster, lower-cost Kling 4.0: 3 to 20 seconds at 720p, with the same kinds of references. It is the one Kling model that runs on free credits.
• MiniMax H3 (minimax-h3) — premium multimodal clips at 2K, with references and audio always on.
• PixVerse V6 Max — 1 to 15 seconds at up to 1080p, from text or a start frame, with sound on by default. It is tuned for film, short drama and advertising.
• Grok Imagine Video 1.5 — animates a start frame for 1 to 15 seconds at up to 1080p. It can end on an exact end frame, make a seamless loop or pass through up to 4 in-between frames.
• Gemini Omni Flash 1.2 — 3 to 15 seconds at any whole second, with native sound, from 360p to 4K. It runs from text, from up to 6 reference images, from exact start and end frames, or as an edit of a clip, and it is priced per second.
• VEO 3.1 Lite and Wan 2.2 Turbo — the cheapest clips, for batches and drafts.

For a comparison of every model, read Choosing a model.

One start frame, three models

The three clips below start from the same image, which Generate Image made, and use the same prompt with sound on. Only the model changes.

• VEO 3.1 Fast kept the golden light of the start frame, pushed in slowly and added the gulls from the prompt.
• Seedance 2.5 stayed closest to the start frame, with the gentlest motion. At 480p it is also the softest picture, which is fine for a draft.
• Gemini Omni Flash left the start frame after about a second and a half: the light turned cooler and the mountains changed.

When a clip has to continue its start frame exactly, watch the first seconds of each take before you build on it, and try a second model when the first one drifts.

PixVerse V6 Max

PixVerse V6 Max is tuned for film, short drama and advertising. It follows the prompt and camera directions closely, with realistic light and materials. Subjects stay consistent across shots, and products and on-screen text look right. It makes 1 to 15 seconds at any whole second (default 5), at 360p, 540p, 720p or 1080p (default 720p). The prompt can be up to 5,000 characters.

• A start frame makes an image-to-video clip. The image is the exact first frame, and the clip takes its shape, so the settings panel hides Aspect Ratio.
• No image makes a text-to-video clip at 16:9 (the default), 4:3, 1:1, 3:4, 9:16, 2:3, 3:2 or 21:9.
• Sound is on unless you turn it off. Untick Generate Audio for a silent clip, which costs less. The sound is ambient audio and effects. Scripted dialogue is not guaranteed, so give a speaking character a voice with text to speech and Lip Sync.
• Seed. The same seed with the same settings gives a similar result.
• No end frame and no references. The editor offers no end frame and no reference image, video or audio input for this model.

PixVerse V6 Max is also offered by Generate Video Pro.

Lengths and resolutions by model

The length and the resolution you can choose depend on the model. The model pages list every value, and the aspect ratios too.

A dash means the model offers no choice of resolution.

Aspect ratios. VEO 3.1, Gemini Omni and LTX 2.3 make 16:9 or 9:16 clips. The Seedance 2 family adds 1:1, 4:3, 3:4, 21:9 and Adaptive, which is its default and matches the connected image. MiniMax H3 and Wan 3.0 also offer Adaptive as their default. A pure text-to-video clip on MiniMax H3 or Wan 3.0 renders 16:9 unless you choose a ratio. HappyHorse 1.1 offers nine ratios for text-to-video and Ref2V clips, including 4:5, 5:4, 21:9 and 9:21; for image-to-video it follows the input image. Vidu Q4 keeps the shape of a start frame used alone; with references it makes 16:9 (the default), 9:16, 1:1, 4:3 or 3:4. Kling 4.0 offers Auto (the default), 16:9, 9:16, 1:1 and 21:9. On Kling 4.0 Keyframes, the ratio applies only when no frame is connected; otherwise the clip follows the start frame's shape. Gemini Omni Flash 1.2 makes 16:9, 9:16 or Auto. With exact frames, Auto picks 16:9 or 9:16, whichever is closer to the start frame. With reference images, it follows the references, and with neither it makes 16:9. PixVerse V6 Max makes eight ratios from text and follows the start frame's shape otherwise. Grok Imagine Video 1.5 makes 16:9, 9:16, 1:1, 4:3 or 3:4, or leaves the ratio to the model.

4K on VEO 3.1. VEO makes the clip at 1080p, then upscales it to 4K in the same run, billed at the 4K price. 4K on Gemini Omni, 4K on Gemini Omni Flash 1.2 and 4K on Vidu Q4 are not available on the free tier.

Start and end frames

Connect an image to Start Frame to open the clip on it. Connect a second image to End Frame to end the clip on it. These models accept an end frame:

• VEO 3.1 Quality, Fast and Lite.
• The Seedance 2 family and Seedance 2.5.
• MiniMax H3 and Hailuo 02 Standard.
• Kling 4.0 Keyframes, Kling 3.0, Kling 3 Omni and Kling 2.5 Turbo Pro. Kling 4.0 Keyframes also accepts an end frame alone.
• Wan 2.7, Wan 3.0 and Wan 3.0 Prime.
• LTX 2.3 Pro and LTX 2.3 Fast.
• Grok Imagine Video 1.5, with a start frame, at 480p or 720p. The same image as both frames makes a seamless loop. See Grok Imagine Video 1.5 frames.
• Gemini Omni Flash 1.2. A start and end frame are exact up to 10 seconds. For a longer clip, or with an end frame alone, the frames join the reference list, and the prompt names them as the opening and closing frame. There the frames guide the clip, but the clip does not open or end on them exactly.

Other models ignore the end frame. Kling 4.0 takes an end frame as a reference image instead, named in the prompt as the closing frame.

Frame fit

A frame that is not the exact size the model renders gets reshaped by the model, and some models do it one frame into the clip. The first frame is your image, and the rest of the clip is slightly taller or wider. Frame fit fixes the size before the frame is sent:

When Nodaro knows the size for your model, resolution and ratio, the panel names it under the setting, for example 720×1280. When your image is more than 5% away from the target shape, for example a square photo in a 9:16 clip, the image is first cropped at the center to the target ratio, then resized. The subject is never squashed. For a combination Nodaro has not measured, the frame is left as it is.

Send frames as

The Seedance 2.0 family (Fast, standard and Mini) zooms a real start frame by about 2% and loses brightness within six frames. Sent as a reference image, the same models keep the look from the first frame. Every other model Nodaro measured reproduces the opening frame more faithfully as a real frame, so Auto keeps them in frame mode. On a model that takes no reference images, this setting is hidden.

Two limits apply. First, a frame never pushes out one of your own reference images: if the frames would exceed the model's image limit, the frame stays a frame. Second, when your request already has reference images, several models cannot keep a real start frame. Seedance, Wan and Vidu Q4 move the frame into the reference list, and MiniMax H3 switches to its reference mode.

Grok Imagine Video 1.5 frames

Grok Imagine Video 1.5 animates a start image. It has no text-to-video mode, so every run needs a start frame. It makes 1 to 15 seconds at any whole second (default 8), at 480p (the default), 720p or 1080p.

• Start and end frame. The start frame is the exact first frame of the clip, and a connected end frame is the exact last frame.
• Seamless loop. Connect the same image to Start Frame and End Frame, and the clip starts and ends on that picture. There is no separate setting.
• In-between frames. The images on Image Refs are up to 4 in-between frames, in the order the settings panel shows. Drag them to reorder. Only images on Image Refs become frames: an asset on Assets never does, and a character adds only its name and description to the prompt.
• Where they land. The frames are spread evenly over the clip, at whole seconds inside it. For example, two frames on an 8-second clip land at 3 and 5 seconds. A short clip has room for fewer frames, as the table below shows. Extra frames are left out, the first ones in your order are kept, and the settings panel shows how many are used.
• A start frame is required. In-between frames alone are refused. An end frame alone is refused by Run, and in a workflow run it becomes the start frame.
• 1080p takes a start frame only. At 1080p the model takes no end frame, no loop and no in-between frames. The editor offers 1080p only while nothing but a start frame is connected. A run that would send more frames at 1080p is refused before any credits are reserved. Choose 720p, or remove the end frame and the in-between frames.
• No other references. The model takes no reference videos, audio or characters. The frames cannot be named in the prompt, so describe in words what happens between them.

Frames add nothing to the price: a start frame alone costs the same as a start frame, an end frame and 4 in-between frames.

References: images, videos and audio

References tell a model what to include without pinning an exact frame. Connect them to Image Refs, Video Refs and Audio Refs. The limits per model are in the table above. Models that do not support references ignore them.

• Seedance 2 family and MiniMax H3. Frames and references can be connected together. With no reference, the frames are exact first and last frames. With any reference connected, the frames join the reference set and the prompt names them as the opening or closing frame. The panel shows the resolved mode and warns you when an input will be dropped.
• Seedance 2.5. Up to 30 images, 10 videos and 10 audio clips. Each reference video must be 2 to 30 seconds long, and all of them together 30 seconds at most. Reference audio can be up to 30 seconds per clip.
• Seedance 2 Fast. Each reference audio clip must be 15.2 seconds or shorter.
• MiniMax H3. Each reference video must be 2 to 15 seconds, and 15 seconds in total. Reference audio is up to 15 seconds per clip, and it must come with an image or video reference; audio alone is not accepted.
• Wan 3.0. Up to 10 images, 5 videos and 5 audio clips. Each video and audio clip is 1 to 15 seconds, with 15 seconds in total per kind. With a reference video, the reference and the output together must stay at 30 seconds or less. A start or end frame joins the reference list, after your own images.
• VEO 3.1 Fast and Lite. Connecting reference images switches VEO to reference mode, with or without a start frame. Reference mode always makes an 8-second clip; VEO's price is the same for 4, 6 and 8 seconds, so the cost does not change. VEO takes at most three images. Your references come first, and a start frame and an end frame follow them. At the limit, the frames are kept and the last references are dropped, so a start frame, an end frame and references leave room for one reference. VEO 3.1 Quality has no reference mode, so its reference inputs are disabled.
• Gemini Omni and Gemini Omni Flash. Up to seven inputs: the start frame and each reference image count as one, a video counts as two. With a start frame, references act as identity references, not frames. References over the limit are dropped, and the node says how many. The start frame is always kept.
• Gemini Omni Flash 1.2. Up to 6 reference images. Name them in the prompt as @image_1, @image_2 and onward. With references connected, the frames follow your references in the same list. The prompt then says in plain words which image is the opening frame, which is the closing frame and which are references for the subject. A frame then guides the clip, usually as its subject, but the clip does not open on it exactly. For an exact opening frame, connect the frames without references. Beside a clip to edit, up to 5 images can be connected, and connected frames count toward those 5. Gemini Omni Flash 1.2 takes no character references and no saved character voices. Use Gemini Omni Flash for those.
• Vidu Q4. Up to 15 images and 3 audio clips, and no reference videos. Reference mode needs a prompt: a run with an empty prompt is refused before any credits are reserved. Reference audio must come with at least one image, and a start frame counts. Each audio clip must be an MP3 file of 3 to 12 seconds. Another format or length is refused before the run starts, so it costs no credits. Export the clip as MP3 first.
• HappyHorse 1.1 Ref2V takes 1 to 9 reference images. Kling 3 Omni takes up to 7.
• Kling 4.0. Up to 10 images, 5 videos, 7 characters and 3 audio clips, with 15 materials in all. See Kling 4.0 references, characters and keyframes.
• Kling 4.0 Flash. Up to 10 images, 5 videos and 7 characters, with 15 materials in all. Reference videos are 2 to 20 seconds together. See Kling 4.0 Flash.

Where the start frame goes. On every model whose frames and reference images share one list, your references come first, in your order. Image number 1 in the prompt is always your first reference. A start frame, and an end frame on models that take one, follow your references, and the prompt names them as the opening and closing frame. At the model's image limit, the frames are kept and the last references are dropped, and the node reports it. This applies to the Seedance 2 family, MiniMax H3, Wan 3.0, Vidu Q4, Kling 4.0, Kling 4.0 Flash, Gemini Omni, VEO 3.1 Fast and Lite, Grok Imagine and HappyHorse 1.1 Ref2V. Until 2026-10-18, Gemini Omni, VEO, Grok Imagine and HappyHorse put the start frame first. Kling 4.0 Keyframes and Grok Imagine Video 1.5 are different: their frames are real frames, and their Image Refs are the in-between frames.

Every Seedance 2 and Seedance 2.5 reference video is re-encoded before it is sent, with the same resolution, frames, length and sound. A clip that Seedance made itself is then edited, instead of coming back as a near-copy of the input. The price does not change.

Most of these length limits are checked before the run starts, so a clip that breaks them costs no credits. Wan 3.0 checks its reference video limits during the run instead, and refunds the reserved credits when a clip is too long. Trim long clips first with Trim Video.

Kling 4.0 references, characters and keyframes

Kling 4.0 comes in two models. Both make 3 to 30 seconds at 720p or 1080p, with native audio always on. Kling 4.0 and Kling 4.0 Keyframes are not available on free credits.

Kling 4.0. With nothing connected, it makes a clip from the prompt alone. Otherwise it takes up to 10 reference images, 5 reference videos, 7 characters and 3 voice clips, with at most 15 materials in all. A connected start or end frame counts as an image. Each reference video is 2 to 30 seconds long, and the reference videos together are at most 30 seconds.

• Frames join the reference images. A start frame or an end frame comes after your own reference images, and the prompt names it as the opening or closing frame.
• Characters become named subjects. A Character node on Assets reaches the model as a subject with the character's name: its portrait and up to 3 more of its images. A character with only one image is sent as a plain reference image.
• The first Audio Refs clip plays as the soundtrack. This is on by default. Kling 4.0 takes reference audio only as voice samples, so Nodaro sends the first clip as one more reference video: a still frame that carries the audio, with that video's sound kept. The result's sound is your exact track, and the people in the scene perform to it. The soundtrack clip is as long as the audio, at least 2 seconds and at most the length of the video; a longer track is cut. In the prompt, @audio_1 names it. The soundtrack takes one of the 5 reference-video places, and the run is priced as a reference-video run. With the soundtrack, Kling also keeps the sound of the reference videos you connect.
• The other audio clips become named voices, in the order they are connected. Kling takes each clip as a sample of a voice and makes its own sound, so a voice clip is not played. To send every clip as a voice sample, turn off Play the first clip as the soundtrack in the settings panel. The input is then called Voice Refs.
• Reference videos must each be 2 to 30 seconds long. A clip outside that range is refused before the run starts, so no credits are spent. A clip larger than 2048 pixels on its long side, or with more pixels than 1920×1080, is scaled down to fit before it is sent, and it keeps its shape. A 4K 16:9 clip is sent as 1920×1080. Its length, and so the price, does not change. By default the clip gets new sound. Through the API, keepOriginalSound: true keeps the reference video's own sound instead.
• Naming materials in the prompt. Write @image_1 for a reference image, @video_1 for a reference video, @Name for a character and @audio_1 for a voice clip. The @ autocomplete writes these for you. If the prompt names a material the run does not carry, Nodaro removes the @ and keeps the word, so the run still starts.
• More than 15 materials is refused before any credits are reserved.

Kling 4.0 Keyframes. The clip moves through a start frame, up to 8 in-between frames and an end frame. The in-between frames are the images on Image Refs, in the order the settings panel shows. Drag them to reorder. An end frame alone is allowed, and with no frame at all the model makes a clip from text. Only images on Image Refs become frames: an asset on Assets never does, and a character adds only its name and description to the prompt. The clip's length limits the number of frames, counting the start, the in-between and the end frames:

More frames than the limit, or more than 8 in-between frames, is refused before any credits are reserved. In the prompt, the frames are @first_frame, @keyframe_1 and onward, and @last_frame. The editor's @image_1 tokens become @keyframe_1 on this model. Kling 4.0 Keyframes takes no reference videos, audio or characters.

Kling 4.0 Flash

Kling 4.0 Flash is the faster, lower-cost Kling 4.0. It makes 3 to 20 seconds at any whole second (default 5), at 720p only. Another resolution renders, and is priced, as 720p. Native audio is always on, with no switch. There is no seed and no end frame, and the prompt can be up to 8,000 characters. Aspect Ratio is Auto by default, which follows your materials and makes 16:9 when there are none. You can also choose 16:9, 9:16, 1:1 or 21:9.

Kling 4.0 Flash is the one Kling model that runs on free credits, as a 3- or 4-second clip at 720p.

The model works in one of two ways, depending on what you connect:

• A start frame on its own is the exact opening frame. No reference image, reference video or character is connected. The clip starts on the frame and takes its shape, so the settings panel hides Aspect Ratio. In the prompt, the frame is @first_frame.
• Anything else runs in reference mode. This covers a prompt alone, reference images, reference videos, characters, or a start frame beside any of them. With nothing connected, the clip comes from the prompt alone. A start frame in reference mode joins the reference images after your own, and the prompt names it as the opening frame.

In reference mode, Kling 4.0 Flash takes up to 10 reference images, 5 reference videos and 7 characters, with at most 15 materials in all. A connected start frame counts as an image. More than 15 is refused before any credits are reserved.

• Characters become named subjects, as on Kling 4.0. A Character node on Assets reaches the model with its portrait and up to 3 more of its images, called @Name in the prompt. A character with only one image is sent as a plain reference image.
• Voice Refs clip N is the voice of character N. Kling 4.0 Flash has no soundtrack clip and no voice on its own. The first clip is the voice of the first connected character, the second clip the voice of the second, and so on. A character needs 2 to 4 images to carry a voice. A clip for a character with one image, or a clip past the last character, is dropped, and the run still starts. Voice clips do not count toward the 15 materials. Write who speaks by the character's name, because a clip cannot be named in the prompt.
• Reference videos must be 2 to 20 seconds long, each and all together. A set outside that range is refused before the run starts, so no credits are spent. As on Kling 4.0, a very large clip is scaled down to fit before it is sent.
• Naming materials in the prompt. Write @image_1 for a reference image, @video_1 for a reference video and @Name for a character. If the prompt names a material the run does not carry, Nodaro removes the @ and keeps the word, so the run still starts.
• No end frame. The settings panel offers none. In a workflow run, an end frame without a start frame becomes the start frame.

Point the prompt at a reference

On reference-capable models, you can tie a phrase in the prompt to one reference, so that it drives the result. Type a token where the phrase belongs. The @ autocomplete in the prompt editor offers them.

For example, with two reference images connected, the prompt circle {image:1:person} wearing {image:2:jacket} for a 360 spin becomes "circle the person from @image_1 wearing the jacket from @image_2 for a 360 spin".

• Numbering. Image numbers count Image Refs first, in order, then the assets on the Assets input. With one reference image and a connected Object, the object is {image:2}. Videos and audio have their own numbers.
• Frames are numbered for you. Start and end frames come after your images, and the prompt names them as the opening or closing frame. You never write a token for a frame.
• A number out of range becomes plain text. {image:5:ghost} on a node with two references becomes "ghost".
• Models without references turn every token into its plain label, so the prompt still reads naturally.

Sound

Many models make an audio track together with the picture. The way you control it depends on the model:

• VEO 3.1 makes sound from the prompt, on by default. Turn off Generate Audio for a silent clip.
• The Seedance 2 family and Wan 3.0 make sound by default. Turn it off with Generate Audio (default on).
• MiniMax H3 always makes sound. With reference audio connected, it syncs the lips to it.
• Gemini Omni makes a soundtrack on every run and takes its direction from the prompt, for example "Sound design: gentle breeze, distant bird chirps". It cannot be switched off, and it does not speak dialogue.
• Gemini Omni Flash 1.2 makes sound with the picture on every run. There is no switch.
• Kling 2.6 makes sound only when Enable Sound is on. Kling 3.0 has Sound Effects. Kling speaks scripted dialogue when you quote the line in the prompt, for example [Anna: warm calm voice]: "good morning". Kling 2.6 speaks English and Chinese; other languages are translated to English by the model.
• Kling 3 Omni includes audio in its price, with a Generate Audio switch.
• Kling 4.0 and Kling 4.0 Keyframes always make sound, included in the price. There is no switch, and they do not speak a character's saved voice. On Kling 4.0, the first clip on Audio Refs plays as the soundtrack unless you turn off Play the first clip as the soundtrack. Kling 4.0 Flash always makes sound too, and its Voice Refs clips become the characters' voices.
• PixVerse V6 Max makes ambient sound and effects unless you untick Generate Audio. A clip with sound costs more. It does not reliably speak scripted dialogue.
• Grok Imagine Video 1.5 makes native sound on every run.

To make the model react to a voice or music, connect it to Audio Refs. To lay your own soundtrack over the finished clip, use a Merge Video & Audio node after Generate Video.

Edit a clip

Seedance 2.5, Gemini Omni and Gemini Omni Flash 1.2 can change a clip you connect to Video Refs.

With Seedance 2.5, the model decides from your prompt whether the run is a normal generation that uses the clip as a reference, or an edit of the clip, for example "remove the sign" or "make it black and white". An edit keeps the source clip's aspect ratio and length. If the model reads your prompt as an edit and your settings ask for another ratio or length, Nodaro submits the run once more with the values the edit needs, and you get the edited clip instead of an error. To ask for an edit from the start, set Use the reference video as to The video to edit. The source clip must be 4 to 30 seconds long. The factory preset Edit Video is built for this kind of run.

With Gemini Omni and Gemini Omni Flash, a connected clip switches the node to video edit. The clip is trimmed to a window of at most 10 seconds, which you set under Source clip trim. The duration follows the clip, and a video edit has a flat price per run.

With Gemini Omni Flash 1.2, a connected clip also switches the node to video edit. Describe the change in the prompt. The result follows the source clip, so Duration is ignored. The clip can be up to 15 seconds long. A longer clip uses its first 15 seconds, and Source clip trim picks another stretch of up to 15 seconds. The price follows the length of the clip. See Credits.

Use the reference video as

On Seedance 2.5, the setting Use the reference video as appears once a clip is connected to Video Refs. It tells the model what the clip is for, so the model does not have to guess it from your prompt.

• The video to edit keeps the clip's own shape and length, so Aspect Ratio and Duration are locked. It applies only when every connected reference video is 4 to 30 seconds long. Otherwise the node runs as with Auto, and a line under the setting says so.
• The video to extend keeps the clip's shape, so Aspect Ratio is locked. Duration stays yours, but Auto becomes 8 seconds, because a continuation needs a set length.
• An extend returns only the new footage. The source clip is not joined to it. For the source and the continuation as one video, use Extend Video.
• Your settings are kept. Edit and extend do not change the stored Aspect Ratio and Duration. Switch back to Auto and they show again. Disconnecting the clip hides the setting and keeps your choice.
• Apply settings restores it. Applying the settings of an earlier Seedance 2.5 result sets this choice back to what that run used.
• The same everywhere. A single run, a workflow run and a published app run send the same choice for the same node and clips.

The price follows the run that is made. See Credits.

Seedance 2.5 Draft mode

On Seedance 2.5, the settings panel has a Draft (480p) switch. A draft is a 480p render of your shot. When you like the take, an Upscale Video node set to Seedance 2.5 HD (from draft) finalizes it at 1080p as the same take, frame for frame. The seed, prompt, references, timing and audio setting stay the same. Use drafts to try a shot several times at a low price, and pay for 1080p only once.

• A draft renders at 480p only. With Draft on, Resolution is locked to 480p. A request or workflow that sends Draft with another resolution renders at 480p, and the change is reported back.
• A draft costs the same as a 480p run, for example 616 credits for 8 seconds, or 374 credits with a reference video.
• The finalize costs extra. Its price comes from the draft's own length. Upscale Video shows the exact price once the draft is wired into it.
• When a draft saves credits. A draft and its finalize together cost more than one direct 1080p run. Drafting first saves credits as soon as a direct 1080p run would need two or more tries.
• Audio carries over. The dialogue of a voiced clip is mixed onto the 1080p render in the same way.
• Drafts are kept for about 7 days. Finalize a draft within that time.
• Use the reference video as works the same. With Draft on, the setting has the same choices and locks. An edit draft keeps the clip's shape and length, and so does its 1080p finalize.

Draft is offered only on Seedance 2.5, and other models ignore it. Video to Video offers the same switch on Seedance 2.5 Edit.

Make a clean loop

Loop trim finds the frame where the clip loops back most cleanly and cuts the clip there. Turn it on in the settings panel:

• Frames to test sets how many frames Nodaro compares, from 4 to 64. The default is 16.
• Quality: Precise cuts at the exact frame with a slight quality loss. Lossless cuts on a keyframe and keeps the file byte for byte.

For a perfect loop, connect the same image to Start Frame and End Frame. Without an end frame, Nodaro still picks the best loop point it finds, but the loop may show a jump. Loop trim also removes the fade that VEO 3.1 adds at the end of a loop.

Loop trim adds a charge to a run from the node's Run button, the API, the SDK or MCP. The price on the Run button does not include it. The charge is one unit for every 5 seconds of the clip or part of 5, plus one unit for every 24 frames it tests or part of 24. Each unit costs 11 credits, the price of one 5-second step of Trim Video. In a workflow run, the trim runs without this charge. If the trim fails, you keep the untrimmed clip and the loop trim credits are refunded.

Keep a character consistent

• Connect assets. Connect a Character Asset, Location Asset, Object/Props Asset or Animal/Creature Asset node to Assets. Nodaro sends their approved pictures and descriptions with the prompt.
• Mention them. Write @maya in the prompt to place the character exactly where the sentence needs it. Give a mention a role, such as @maya:1:clothes, to say what to take from the reference. See Reference roles.
• Check what is sent. The Injected references list in the settings panel shows every reference the model will receive. Drag to reorder, or click × to remove one.
• Use a board. A reference board or a consistency grid from Generate Image keeps a face stable across clips. The Scene Recipes presets are built to read these boards. See Reference boards.

Read Consistent characters for the full method.

Presets

Open Presets on the node for ready-made setups. Each preset chooses a fitting model, aspect ratio and duration, fills the prompt with {placeholder || default} slots, and adds a tuned negative prompt against warping, flicker and morphing. You can run a preset unedited: every slot falls back to its default.

To chain several Scene Recipes into one short, join the clips with the Seamless Join (One-Shot) preset of Combine Videos. Read Presets to save and share your own.

Credits

The cost badge on the node shows the price before you run. The price depends on the model, the duration and the resolution. On some models it also depends on sound and references.

• Flat price per clip. VEO 3.1 costs the same for 4, 6 or 8 seconds. VEO 3.1 Lite costs 83 credits at 720p and 99 at 1080p. VEO 3.1 Fast costs 165 at 720p and 187 at 1080p. VEO 3.1 Quality costs 1,100 for a 1080p clip.
• Price per duration. Gemini Omni and Gemini Omni Flash have a price per length and resolution. Gemini Omni Flash costs 176 credits for 4 seconds and 352 for 10 seconds at 720p or 1080p. A video edit has one flat price per run: 462 credits on Gemini Omni Flash and 660 on Gemini Omni, at 720p or 1080p.
• Gemini Omni Flash 1.2 is priced per second. Each whole second from 3 to 15 has its own price at 360p, 720p, 1080p and 4K. A video edit is charged for the clip it sends: its seconds after any trim, at most 15, rounded up to a whole second, with a minimum of 3 seconds. When the clip cannot be measured, the run holds the 15-second price, then charges the length delivered.
• Price per second. Seedance 2.5, MiniMax H3, Wan 3.0, Grok Imagine Video 1.5, PixVerse V6 Max and HappyHorse 1.1 are priced by the second. For example, Wan 3.0 costs 220 credits for 5 seconds at 720p and 704 for 8 seconds at 1080p. Vidu Q4 is priced by the second too, with one price for both of its modes, with or without sound, and references add nothing. Without a reference video, the Seedance 2 family bills in tiers of 4, 8, 12 and 15 seconds, and a length between two tiers bills as the next one, so 6 seconds bills as 8.
• Kling 4.0 is priced by the second. Kling 4.0 and Kling 4.0 Keyframes share one price per second of length at each resolution. Sound is included, and reference images, characters, voice clips and frames add nothing. A reference video on Kling 4.0 bills the output plus the longer of the output and the reference videos, at a reference rate. For example, a 5-second clip with a 4-second reference video bills as 10 seconds. A soundtrack clip counts as a reference video. It is never longer than the video, so a soundtrack run with no other reference video bills twice the video's seconds at the reference rate.
• Kling 4.0 Flash is priced by the second. Each length from 3 to 20 seconds has its own price at 720p, whether or not a start frame is connected. Sound is included, and reference images, characters, voice clips and the start frame add nothing. A run with no length set is priced as 5 seconds. A reference video bills the output plus the longer of the output and the reference videos, at a reference rate, as on Kling 4.0.
• Grok Imagine Video 1.5 frames add nothing. Each length has its own price at 480p, 720p and 1080p. An end frame, a loop and up to 4 in-between frames cost the same as a start frame alone. A request without a length is priced as 8 seconds, and one without a resolution as 480p.
• PixVerse V6 Max has a silent price and a price with sound. Each length and resolution has both, and the price with sound is higher. Sound is on unless you untick Generate Audio. The price is the same with or without a start frame.
• Reference videos make Seedance cheaper. On the Seedance 2 family and Seedance 2.5, connecting a reference video switches to a lower price per second. Seedance 2 at 1080p costs 2,244 credits for 8 seconds, or 1,364 for 8 seconds of output with a reference video.
• Sound can cost more. Kling 3.0 costs 748 credits for 10 seconds without sound and 1,100 with sound. On Kling 3.0, sound is on unless you turn it off.
• Reference videos count their own length. On the Seedance 2 family and MiniMax H3, the price covers the seconds of each reference video plus the seconds of output. MiniMax H3 at 2K costs 803 credits for 8 seconds of output, and each second of a reference video adds the same price per second. On Wan 3.0, a reference video adds nothing.
• Extra images on MiniMax H3. Each input image beyond the first five adds a small charge. Six seconds at 2K cost 605 credits with up to five images, plus that charge for each image after the fifth.
• Auto length is settled after the run. With Duration set to Auto, Nodaro reserves the price of the model's longest clip: 30 seconds on Seedance 2.5 and 15 seconds on the other Seedance 2 models. When the clip is ready, you pay for the length you got and the rest is refunded. Your balance must cover the reservation for the run to start.
• Edits are settled after the run too. A Seedance 2.5 run with a reference video is reserved for the longer of your duration and the clip, then settled to the length delivered.
• Auto with a reference video is usually an edit of the first clip. On the Seedance 2 family, the edit keeps the length of the clip it edits, but the model may edit another clip. When every reference clip is measured, Nodaro reserves the output at the longest clip's length, rounded up to a whole second, plus one second. The output is never reserved above the model's longest clip, and you pay for the length delivered. Every connected clip still counts on the input side. A clip that cannot be measured is reserved at the worst case: the model's longest clip for the output, and the longest clip the model accepts for the input.
• Use the reference video as changes the reservation. On Seedance 2.5, The video to edit is priced as an automatic-length run with a reference video from the start, and your Duration is not used. A reference with Duration on Auto is reserved at the longest output the model can make, then charged what it makes. The video to extend is priced at the length it runs at.
• An unmeasurable reference clip is measured again. On the Seedance 2 family, Kling 4.0 and Kling 4.0 Flash, a reference clip that cannot be measured is reserved at the model's longest accepted clip. When the run is done, Nodaro measures the clip again. If it can be read then, you are charged its real length, and the rest is refunded.
• The price on the node is the exact reservation. On the Seedance 2 family, the node's price and its Run button come from the server, which measures the connected reference videos and prices them the way the run does. While a reference video has not been made yet, the node shows a range, from the shortest clip the model accepts to its worst case, and the Run button shows the top of the range. No figure is shown while the clips are being measured.
• Workflow and app estimates price reference videos the same way. A workflow run's estimate and a published app's listed price measure an uploaded reference clip. A clip they cannot know, such as one a node in the run makes or one the app's user uploads, is priced at the model's worst case. A quote is never below what the run reserves.
• Connected settings are priced as they run. With a Provider or Duration node on the Settings input, the price uses the connected model and the fitted length. See Settings input.
• Loop trim is extra. On a run from the node, the API, the SDK or MCP, it adds a charge that the price on the Run button does not include. See Make a clean loop.

Every model's page lists its exact prices. See Credits for how reservations and refunds work.

Steps after a video of automatic length

Some clips come out at a length that is not the Duration stored on the node: a Seedance 2 family clip on Auto, a Seedance 2.5 edit, and a Seedance 2.5 extend on Auto. A Seedance 2.5 run with A reference or The video to extend records its length the same way. Such a clip records the length it really is when it is made, never above the longest that run could make.

• The next step is charged by that length. Trim Video, Loop Video and Combine Videos are charged by the recorded length, in a single-node run, a workflow run and a published app run alike.
• Quotes count the longest length until the clip exists. The Run button, the workflow total, an app's price and the credit check before a run count such a clip at its longest possible length. Once the clip exists, they use its recorded length. A quote can be above the charge, never below it. For example, Seedance 2.5 on Auto into Combine Videos is quoted as a 30-second clip, then charged for the length it came out at.
• Video Analysis and AI Audit follow it too. In a workflow run, they are priced by the recorded length.
• Everything else is unchanged. A run at a set Duration with no task type hands on its Duration. Steps that measure their own input, such as Video SFX and Mask Composite, keep their own measurement.

When a setting is not supported

A saved workflow can carry a resolution, aspect ratio or duration that the current model does not offer, for example after you switch models. Nodaro does not fail the run. It corrects the value to one the model accepts, and you pay for the corrected value, because that is also what the model makes.

• The nearest value wins. A 4K request on a model that stops at 1080p renders 1080p. A portrait 9:21 becomes 9:16, never landscape.
• An empty resolution is sent at the resolution it is priced at. The value used is recorded on the run.
• Adaptive and Auto stay as they are. They tell the model to match the input.
• Duration stays as you set it, except on LTX 2.3, which moves to its nearest length. A 7-second request on LTX 2.3 renders and bills 6 seconds.
• A connected Duration node is fitted. A length that comes from the Settings input always runs as the nearest length the model makes, and you pay for that length.
• MiniMax H3, Wan 3.0, Vidu Q4, Kling 4.0, PixVerse V6 Max and Grok Imagine Video 1.5 render a fixed default for an unsupported resolution: 2K on MiniMax H3, 480p on Grok Imagine Video 1.5, and 720p on Wan 3.0, Vidu Q4, PixVerse V6 Max and every Kling 4.0 model, Kling 4.0 Flash included. The price follows what they render.

When a model blocks the prompt

Some models check a finished clip against their content policies, for example for resemblance to protected film and TV content, and reject it. When that happens, Nodaro rewrites the prompt once. The rewrite keeps the same subjects, camera language and mood, and softens what triggered the rejection. Nodaro then runs the clip again at no extra credit cost. If the second run is also rejected, the run fails with the model's own reason.

Older Text to Video and Image to Video nodes

Generate Video replaces the older Text to Video and Image to Video nodes. A saved workflow that contains them opens with Generate Video nodes in their place. The connections move to the matching inputs, and the models, settings and prices stay the same. You do not need to change anything.

Tips

• Put the important words first. Text past your model's prompt limit is cut off, and Kling 2.6 allows only 1,000 characters.
• Draft cheap, finish sharp. Iterate on Seedance 2.0 Fast or VEO 3.1 Lite at a low resolution, then switch the model or the resolution for the final clip. On Seedance 2.5, turn on Draft (480p) and finalize the take you like with Seedance 2.5 HD.
• Describe motion, not the picture. With a start frame, the image already shows the scene. Spend the prompt on what moves and on the camera.
• Describe a camera move, do not just name it. Say what the camera physically does, what stays still and where the shot ends. Camera Motion Lab shows, over 147 test clips, why a bare term such as "orbit" or "dolly" goes wrong so often.
• Put the look in pickers. Pickers such as Camera Motion and Lighting add tested wording, and you can change them without rewriting the prompt.
• Drive many shots from one node. Wire a List of shot prompts into one Generate Video node instead of copying the node, so the model is set once for every shot.
• Keep several video nodes on one model. Connect one Provider node to the Settings input of each of them, and switch the model in one place.
• Connect only what the model uses. The settings panel hides controls the model does not support, but values you set for another model stay in the node.

Troubleshooting

Run is disabled and the tooltip names the model. The model has no text-to-video mode. Connect an image to Start Frame, or choose a model that makes clips from text.

Run is disabled and a chip on the node is red. The Provider node on the Settings input holds a model this node cannot run, such as an image model. Choose one of the node's video models in the Provider node. A workflow run stops at this node before anything is charged, and its error names the Provider node.

The clip's length or shape differs from the connected setting. The model does not make that length or ratio, so the node runs the nearest one. The chips on the node show the values it runs with. See Settings input.

The clip jumps slightly after the first frame. The model reshaped your start frame. Set Frame fit back to Match output size.

On Seedance 2.0, the first frames look zoomed or darker. Set Send frames as to Reference image, or leave it on Auto, which does this on the Seedance 2.0 family.

The node says an input will be dropped. You connected more references than the model accepts, or you combined inputs the model cannot use together, such as a VEO end frame with reference images. Remove inputs, or choose a model with a higher limit.

A reference clip is rejected before the run. A reference video or audio clip is outside the model's length limits. Trim it with Trim Video or Trim Audio, then run again. No credits are spent.

A Wan 3.0 run fails after it starts. Wan 3.0 checks its reference video limits while the run is in progress. The reserved credits are refunded. Shorten the reference videos and run again.

The Seedance 2.5 result has a different length or shape than you set. The model treated your prompt as an edit of the reference video, which keeps the clip's own ratio and length. To decide this yourself, set Use the reference video as.

Kling 4.0 refuses the run before it starts. The run carries more than 15 materials, or its reference videos add up to more than 30 seconds (20 seconds on Kling 4.0 Flash). On Kling 4.0 Keyframes, the run can also have more frames than the clip's length allows. Remove inputs, use shorter reference videos, or make the clip longer. No credits are spent.

Grok Imagine Video 1.5 refuses the run at 1080p. At 1080p the model takes a start frame only, with no end frame, loop or in-between frames. Choose 720p, or remove the end frame and the in-between frames. Nothing is charged.

4K is rejected on Gemini Omni or Gemini Omni Flash 1.2. 4K is not available on the free tier. Choose 720p or 1080p, or upgrade your plan.

Some models, lengths or resolutions are locked. Until your account buys anything, video on free credits is limited to clips of up to 4 seconds at the model's lowest resolution. Seedance 2.5 and every Kling model except Kling 4.0 Flash are not available. The locked options say why under them. The default settings, Seedance 2 Fast at 4 seconds, run as they are. Any credit pack or plan lifts the limits.

From the API

Every setting above is also available to code and to AI assistants through POST /v1/generate-video. The older route POST /v1/text-to-video still works and behaves the same way.

• Direction by id. Send a direction object of picker ids, for example { "cameraMotion": "dolly-in", "timeOfDay": "dawn" }, and Nodaro writes the matching wording into the prompt. Valid ids come from GET /v1/picker-catalogs. See Picker catalogs.
• Frames. frameFit and frameDelivery set Frame fit and Send frames as.
• Corrections. Corrected values come back in an adjustments array.
• Drafts. draft: true with provider set to seedance-2-5 makes a 480p draft. Finalize it with POST /v1/video-upscale and { "provider": "seedance-2-5-hd", "sourceJobId": "<the draft's job id>" }. To read the finalize price first, without starting a run, call POST /v1/credits/seedance-hd-estimate with the same sourceJobId.
• Kling 4.0. Send provider: "kling-4" with referenceImageUrls, referenceVideoUrls, referenceAudioUrls and characterReferences. Each character reference is a named subject: its imageUrl, bodyImageUrl and up to 3 extraImageUrls are its 2 to 4 images, and its name is how the prompt calls it. A voice on a character reference is refused; send voice clips as referenceAudioUrls. The first clip plays as the soundtrack; audioSoundtrack: false keeps every clip a voice sample.
• Kling 4.0 Flash. Send provider: "kling-4-flash" with the same fields. Clip N of referenceAudioUrls is the voice of character reference N, and a voice on a character reference is refused. keepOriginalSound: true keeps a reference video's own sound. A request with only an end frame and no references is refused.
• Your own soundtrack. audioUrl still lays a track over the result, but it is deprecated. Send the clip and the audio to Merge Video & Audio instead. For kling-4-keyframes, send the in-between frames as referenceImageUrls, in order.
• Gemini Omni Flash 1.2. Send provider: "gemini-omni-flash-1-2" with a whole-second duration from 3 to 15, a resolution of 360p, 720p, 1080p or 4k, and an aspectRatio of 16:9, 9:16 or auto. A source video is edited, and videoTrimStart and videoTrimEnd pick a stretch of up to 15 seconds. A seed is sent only when it is a whole number from 10000 to 99999. Without resolution the clip is priced at 720p, and without duration at 8 seconds. The model takes no characterReferences.
• Seedance 2.5 reference task type. referenceTaskType (auto, reference, edit or extend) sets Use the reference video as. Through MCP, it is reference_task_type on generate_video.
• PixVerse V6 Max. Send provider: "pixverse-v6-max". Sound is on unless the request sends sound: false; without sound, generateAudio: false does the same. A seed is a whole number from 0 to 2147483647. Without resolution the clip renders at 720p, and without duration it makes 5 seconds.
• Grok Imagine Video 1.5. Send provider: "grok-imagine-video-1.5" with a start frame. endFrameUrl is the exact last frame, and referenceImageUrls are up to 4 in-between frames, in order. At 1080p, a request with an end frame or in-between frames is refused with 400 and the code start_frame_only_resolution.
• Defaults. A request without provider runs on Seedance 2 Fast. If duration is also missing, it makes a 4-second clip.
• Character voice. On models that can speak dialogue, such as VEO 3.1, Kling 2.6, Kling 3.0, Kling 3 Omni, the Seedance 2 family and MiniMax H3, you can pass characterVoices and dialogue. The clip then speaks in the character's saved voice, for an extra 44 credits on VEO and Kling, or 5–28 credits per 1,000 characters on Seedance and MiniMax H3. On other models the voice is ignored, the clip still generates and no extra credits are charged.

See Run a single node and the MCP tools.

Frequently asked questions

Which model should I choose in Generate Video?

New nodes start on Seedance 2 Fast, a low-cost model with sound. Use VEO 3.1 Fast for a good balance of price and quality with native audio, Kling 3.0 or VEO 3.1 Quality for premium cinematic shots, and Seedance 2 or Seedance 2.5 when references drive the shot.

How do I switch between text to video and image to video?

You do not switch anything. With only a prompt, the node makes a text-to-video clip. Connect an image to Start Frame and the same node animates that image. Connect an End Frame too and the clip moves from one image to the other.

How many credits does a Generate Video clip cost?

The price depends on the model, the duration and the resolution, and on some models on sound and references. The cost badge on the node shows the price before you run. For example, VEO 3.1 Lite costs 83 credits for a 720p clip, and Seedance 2 costs 902 credits for 8 seconds at 720p.

How long can a video clip be?

It depends on the model. VEO 3.1 makes 4, 6 or 8 seconds, Kling 3.0 and Seedance 2 up to 15 seconds, and Seedance 2.5, Wan 3.0 and Kling 4.0 up to 30 seconds in one run. For longer videos, use Generate Video Pro, which stitches several segments into one clip.

Does Generate Video make sound?

Many models make an audio track with the picture, including VEO 3.1, the Seedance 2 family, Kling 3.0, Gemini Omni, MiniMax H3, Wan 3.0 and PixVerse V6 Max. Turn it off with Generate Audio where the model allows it. To lay your own track over the finished clip, use a Merge Video & Audio node.

What happened to the Text to Video and Image to Video nodes?

They are now part of Generate Video. Saved workflows that contain them open with Generate Video nodes in their place, with the same connections, settings and prices.

Related

Generate Video Pro

Make one long AI video from a prompt or a script. Generate Video Pro splits it into model-sized segments, renders them in order and stitches one clip.

Generate Image

Create an image from a text prompt with one of Nodaro's image models, keep characters consistent with references, and edit results in place with a mask.

Extend Video

Continue a video with new footage from your prompt, using Seedance 2.5, Seedance 2.0, Kling 4, LTX 2.3 Pro, VEO or Runway Extend. Chain runs for long shots.

Reference roles

Give each reference a role such as face, outfit, background or style, lock a subject's identity when you need it, and mention references by name in the prompt.

Video models

Every video model you can run in Nodaro, with what each one does and what it costs in credits.