Pricing

The following table lists the pricing of the current Partner Nodes. All prices are in credits. 211 credits = 1 USD

​

Beeble

​

BFL

​

Anthropic

Anthropic partner nodes bill the same underlying USD usage as the public Anthropic API. Credits = USD × 211 (see page header). Token prices below are per million tokens. Cache pricing multipliers match Anthropic’s prompt caching documentation.

​

Chat (ClaudeNode)

Models: per 1M tokens. Optional images add tokens billed at the same model input rates.

​

Bria

​

ByteDance

Seedance 2.0 token calculation: Output tokens = (Input video duration + Output video duration) × Output video length × Output video width × Output video frame rate ÷ 1024. When input is text, image, or audio, input duration = 0. Seedance 2.0 outputs at 24fps. For video-to-video inputs, a minimum token usage limit applies. Per‑K token credits depend on output resolution: 1080P is 2.5× the 720P rate for the same model and video_type.

​

ElevenLabs

​

Freepik

​

Google

​

HappyHorse

​

Hitpaw

​

Video Enhancement

​

Photo Enhancement

​

Ideogram

​

Krea

​

Kling

​

Lightricks

​

Luma

​

Meshy

​

Minimax

​

Moonvalley

​

OpenAI

OpenAI partner nodes bill the same underlying USD usage as the public API. Credits = USD × 211 (see page header). Text-model USD rates below match OpenAI’s API pricing Standard tier (per 1M tokens) as published on that page; image-token multipliers match comfy_api_nodes/nodes_openai.py. DALL·E and Sora fixed math matches the nodes’ price_badge expressions. Final charges follow metered usage (tokens, duration, and any cache hits).

​

DALL·E 2 (OpenAIDalle2)

Per generated image (multiply by n when you request more than one).

​

DALL·E 3 (OpenAIDalle3)

​

GPT Image 1 / 1.5 / 2 (OpenAIGPTImage1)

Token-based from API usage. Credits = USD × 211 per 1M tokens (rates from the node’s calculate_tokens_price_image_1 / calculate_tokens_price_image_1_5 helpers). The UI shows approximate USD ranges per run by quality (low / medium / high); actual cost depends on token counts.

​

Chat (OpenAIChatNode)

Models selectable in the node: gpt-5.5, gpt-5.5-pro, gpt-5, gpt-5-mini, gpt-5-nano, gpt-4.1, gpt-4.1-mini, gpt-4.1-nano, o3, o4-mini, o1, o1-pro. Standard tier, per 1M tokens (cached column is OpenAI’s cached-input rate where listed). For models with both short- and long-context pricing, the long-context rate applies once the request exceeds the short-context window. Optional images and files add tokens billed at the same model input rates.

​

Sora video (OpenAIVideoSora2)

USD is (per-second rate) × (duration in seconds); duration can be 4, 8, or 12. Credits = USD × 211. Example: sora-2, 1280x720, 8 seconds → about 168.8 credits ($0.10 × 8 × 211).

​

OpenRouter

​

OpenRouter LLM (OpenRouterLLMNode)

Per 1M tokens.

​

Pika

​

Pixverse

​

Quiver

​

Recraft

​

Reve

Prices shown are base prices (test_time_scaling=1, no postprocessing). Actual cost may be higher with upscaling or background removal enabled.

​

Rodin

​

Hyper3D Rodin (Gen-2)

​

Hyper3D Rodin Gen-2.5 (Rodin3D_Gen25_Image / Rodin3D_Gen25_Text)

​

Runway

​

Sonilo

​

Stability AI

​

Tencent

​

Topaz

​

Image Enhancement

​

Video Enhancement

​

Tripo

​

Vidu

​

Video Generation — Q3

​

Video Generation — Q2

Off-peak price is half the normal price. Decimals round up.

​

Video Generation — Q1

​

Video Generation — Vidu 2.0

​

Image Generation

​

Video Extension

​

Other Features

​

WAN

​

Wavespeed

​

xAI

​

Other (Cloud)