Model pricing
Prices below are per generation, in US dollars, as charged by the providers at the time of writing. Each range goes from the cheapest setting to the most expensive, resolution and length being what moves it. The node pages have the full breakdown and what each model is good at.
In the app you don’t have to look anything up: the price of a run is written under Generate, before you press it, along with how many more runs your provider balance pays for.
Image
The Image node makes still images from a prompt, with or without reference images.
| Model | Resolution | Price per image |
|---|---|---|
| Nano Banana Pro | 1K · 2K · 4K | $0.14–0.24 |
| Nano Banana 2 | 1K · 2K · 4K | $0.045–0.14 |
| Nano Banana | HD | $0.038 |
| GPT Image 2.5 Sunburst | 1K · 2K · 4K | $0.01–1.00 |
| GPT Image 2.5 Flare | 1K · 2K · 4K | $0.01–1.00 |
| GPT Image 2 | 1K · 2K · 4K | $0.01–0.72 |
| Midjourney | Standard · HD | $0.10–0.15 for 4 images, WaveSpeed only |
| Qwen Image 3.0 | 1K · 2K | $0.03–0.075 |
| Seedream 5.0 Pro | 1K · 2K | $0.045–0.09 |
| Flux 2 Max | HD | $0.07 |
| Z-Image | HD | $0.005 |
| Ideogram 4.0 | 1K · 2K | $0.025–0.20 |
| ERNIE | HD | $0.03 |
Video
The Video node makes clips from a prompt, from images, or from a clip you already have. Length and resolution drive the price more than anything else.
| Model | Resolution | Duration | Price per clip |
|---|---|---|---|
| Seedance 2.5 | 480p · 720p · 1080p · 4K | 4–30s | $0.72–54 |
| Seedance 2.0 | 480p · 720p · 1080p | 5 · 10 · 15s | $0.50–9.00 |
| Kling O3 | 720p · 1080p · 4K | 4–15s | $0.34–6.30 |
| Kling V3 | 720p · 1080p · 4K | 4–15s | $0.34–6.30 |
| FLUX 3 | 720p · 1080p | 5–20s | $0.85–5.80 |
| Veo 3.1 | 720p · 1080p · 4K | 4 · 6 · 8s | $0.40–1.60 |
| Gemini Omni | 720p | 3–10s | $0.39–1.30 |
| LTX 2.3 | 480p · 720p · 1080p | 5 · 10s | $0.10–0.40 |
3D
The 3D node turns a picture, or a prompt, into an object you can turn.
| Model | Polygons | Price per object |
|---|---|---|
| Hunyuan3D V3 | 40k – 1.5M | about $0.38 |
| Hunyuan 3D 3.1 Pro | 3k – 1.5M | Runware only, see the node page |
Avatar
The Avatar node makes a portrait talk from a voice recording. The price follows the length of the audio: the Kling models bill by the second, with a five-second minimum.
| Model | Max length | Price per clip |
|---|---|---|
| Kling AI Avatar 2.0 Pro | 5 min | $0.56–33.60 |
| Kling AI Avatar 2.0 Standard | 5 min | $0.28–16.80 |
| HeyGen Avatar IV | 2 min | Runware only, see the node page |
Upscale
The Upscale node enlarges an image. The target size, in megapixels, is what drives the price.
| Model | Max size | Price per image |
|---|---|---|
| Clarity AI Crystal Upscaler | 1500 MP | $0.05–24.00 |
| Clarity AI Creative Upscaler | 64 MP | $0.05–1.60 |
| Clarity Upscaler | 64 MP | $0.12–1.92 |
Prompt
The Prompt node writes prompts for the generators. It costs a fraction of a generation: around a cent a reply, or five cents when the model has to look at connected images. The model you pick here, ChatGPT 5.5, Qwen 3.6 Plus or Qwen3.5 27B, matters far less to your bill than the one on the generator it feeds.
Prices are indicative and set by the providers, who revise them: see wavespeed.ai or runware.ai for the current list. The node shows the exact price of a generation before you run it.