NS WaveSpeed Wan 2.5 Image to Video Fast
Iterate on the still, save the full-price run for the good take
- client
- video_url
NSWaveSpeedWan25ImageToVideoFast from symbiotica-ai/comfyui-nodes is the last node in the pack's Wan 2.5 lineup, and it's the one that makes the rest affordable. It's the Fast tier of the i2v node - same hosted Wan 2.5 engine, cheaper per call - with one structural difference worth knowing: no 480p. Where the standard i2v node offers 480p/720p/1080p, Fast starts at 720p and goes to 1080p. Fast isn't "lower resolution," it's "fewer steps for the same resolution." You're paying less per render because the render is quicker, not because the canvas is smaller.
How it works
Your image (URL string from Upload Image node) and a motion prompt go to WaveSpeed's Wan 2.5 fast-i2v endpoint, and a video_url comes back. The inputs mirror the standard node - audio conditioning (URL to a 3–30s, ≤15MB wav/mp3), duration (5 or 10s), negative_prompt, enable_prompt_expansion, seed (-1 = random) - minus the 480p resolution rung. enable_sync_mode defaults off: submit, poll every second, block until done, Cancel honored.
The intended workflow is the same as every Fast tier in this pack: use the cheap node to find the right motion - try five prompts against the same still - then run the winner through the standard node for the final render. On a metered API, that split is where you keep the session cheap and still end with the good take.
The inputs that matter
image- the still to animate, as a URL.prompt- the motion.resolution- 720p or 1080p (no 480p here).audio(optional) - URL to audio conditioning.duration- 5 or 10 seconds.client- theWAVESPEED_AI_API_CLIENTsocket from the NS WaveSpeed Client node.
Output: video_url (STRING) - browser-preview or feed a download-from-URL node.
Install and keys
# ComfyUI Manager: search "Symbiotica", or:
cd ComfyUI/custom_nodes
git clone https://github.com/symbiotica-ai/comfyui-nodes.git symbiotica
pip install -r symbiotica/requirements.txt
Restart, add the NS WaveSpeed Client node with your wavespeed.ai key (widget, Settings → Symbiotica, or WAVESPEED_API_KEY), wire client in. No model files.
Common issues
Fast is for finding, standard is for finishing. The classic mistake is leaving the Fast node in place for the final render out of habit and shipping a clip a notch below the standard tier. Make the swap explicit.
Need a cheap 480p test? That's on the standard i2v node, not here. If your iteration loop doesn't need 720p+, the standard node at 480p might actually be the cheaper iteration tier than Fast at 720p - read the pricing before you assume.
Hosted-only, as ever. Wan 2.5 has no open weights; local means Wan 2.2 (wan-video.md). Every call here is metered and your image leaves the machine (external-api-nodes.md).
Inputs (10)
| Name | Type | Default | Description |
|---|---|---|---|
| client | WAVESPEED_AI_API_CLIENT | — | |
| image | STRING | Image URL to animate (connect from Upload Image node) | |
| prompt | STRING | Text description for video animation | |
| resolution | COMBO | 720p | Video output resolution |
| negative_promptopt | STRING | Describe what you don't want in the video | |
| audioopt | STRING | Audio URL to guide video generation (3-30 seconds, wav/mp3, ≤15MB) | |
| durationopt | COMBO | 5 | Video duration in seconds |
| enable_prompt_expansionopt | BOOLEAN | false | Automatically expand and enhance the prompt |
| seedopt | INT | -1-1–2147483647 | Random seed for reproducible results. -1 for random seed |
| enable_sync_modeopt | BOOLEAN | false | Wait for generation to complete before returning |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| video_url | STRING | — |