NS WaveSpeed Wan 2.5 Text to Video Fast

The Fast tier of hosted Wan 2.5 text-to-video, NSWaveSpeedWan25TextToVideoFast from symbiotica-ai/comfyui-nodes, is what you use while you're still finding the prompt. Same model family as the standard node, same hosted pipeline, lower per-call cost and shorter wait - with the tradeoff that Fast renders are the iteration tier, not the ship-it tier. If you've used the Veo or Wan Fast nodes in this pack, you already know the playbook; this is that playbook pointed at Wan 2.5, the model you can't download.

How it works

Prompt and size in, video_url out - a hosted Wan 2.5 render on WaveSpeed. The differences from the standard node are in the size menu and the price. Fast offers four sizes - 1280*720, 720*1280, 1920*1080, 1080*1920 - the same 1080p ceiling, minus the 480p budget options. Everything else carries over: audio conditioning (a hosted URL, 3–30s, ≤15MB), duration (5 or 10s), negative_prompt, enable_prompt_expansion, and seed (-1 = random). enable_sync_mode defaults off - submit, poll every second, block until done, Cancel honored.

The workflow that makes Fast worth it: run your prompt experiments here, at the cheaper tier, until a take is right - then switch the same prompt to the standard node for the final render. On a metered API that pattern is the difference between "exploring a prompt is expensive" and "exploring a prompt is a rounding error."

The inputs that matter

• prompt - the scene and motion.
• size - 1280720, 7201280, 19201080, 10801920.
• audio (optional) - URL to audio conditioning.
• duration - 5 or 10 seconds.
• enable_prompt_expansion - auto-enrich.
• client - the WAVESPEED_AI_API_CLIENT socket 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

Don't ship from Fast by muscle memory. It's cheap for a reason; the standard node is the finishing tier. Keep the swap explicit in your workflow so you don't pay iteration prices for a final render - or worse, ship an iteration-grade clip as if it were final.

480p isn't here. If you specifically want the budget 832*480 option, that's the standard node's menu. Fast's floor is 720p.

Wan 2.5 remains hosted-only. Whatever tier you pick, there's no local 2.5 - the newest open Wan is 2.2 (wan-video.md), and every call here is metered and leaves your machine (external-api-nodes.md).