NVIDIA modelsvideo generation api

nvidia/cosmos-3-super/image-to-video

Name: Cosmos 3 Super Image to Video
Author: NVIDIA

Cosmos 3 Super is NVIDIA's image-to-video AI model. Bring static images to life with fluid animation, consistent character motion, and professional-grade video output.

Input

Prompt *

This field is required.

Image Url *

This field is required.

Guidance scale (CFG)

020

Num Inference Steps

150

Agentic Samples Per Iteration

Agentic Early Stop

Enable Agentic Generation

Num Frames

5189

Frames Per Second

460

Agentic Max Iterations

Aspect Ratio

Seed

Idle

Example output — click Run to generate your own

API README

PRICED DRAFT — MANUAL REVIEW REQUIRED source: fal.ai pricing metadata source_url: https://fal.ai/models/nvidia/cosmos-3-super/image-to-video exact_model_mode: nvidia/cosmos-3-super/image-to-video original_unit: Your request will cost $0.05 per second of generated video, rounded up. Agentic generation is billed for each candidate video generated method: fal per-started-generated-second and per-generated-candidate rule result_base_price: 0.4 result_price_formula: Math.ceil((usage.total_seconds ?? ((params.num_frames ?? 189) / (params.frames_per_second ?? 24)))) * (usage.output_count ?? ((params.enable_agentic_generation ?? false) ? (params.agentic_samples_per_iteration ?? 2) * (params.agentic_max_iterations ?? 2) : 1)) * 0.05 review_status: pending; model and provider remain disabled checked_at: 2026-08-16

More Models by NVIDIA

nvidia/nemotron-3-ultra-550b-a55bNVIDIA Nemotron 3 Ultra is an open frontier-reasoning and orchestration model from NVIDIA, with 55B active parameters out of 550B total (MoE). Built on a hybrid Transformer-Mamba mixture-of-experts ar...$$nvidia/nemotron-3-nano-30b-a3bNVIDIA Nemotron 3 Nano 30B A3B is a small language MoE model with highest compute efficiency and accuracy for developers to build specialized agentic AI systems. The model is fully open with open-wei...$nvidia/nemotron-3-super-120b-a12bNVIDIA Nemotron 3 Super is a 120B-parameter open hybrid MoE model, activating just 12B parameters for maximum compute efficiency and accuracy in complex multi-agent applications. Built on a hybrid Mam...$nvidia/cosmos-3-super/text-to-imageCosmos 3 Super by NVIDIA - generate stunning images from text prompts with state-of-the-art AI. Supports multiple aspect ratios, styles, and high-resolution output for creative and commercial use.Pay per use nvidia/nemotron-nano-9b-v2NVIDIA-Nemotron-Nano-9B-v2 is a large language model (LLM) trained from scratch by NVIDIA, and designed as a unified model for both reasoning and non-reasoning tasks. It responds to user queries and t...$nvidia/llama-3.3-nemotron-super-49b-v1.5Llama-3.3-Nemotron-Super-49B-v1.5 is a 49B-parameter, English-centric reasoning/chat model derived from Meta’s Llama-3.3-70B-Instruct with a 128K context. It’s post-trained for agentic workflows (RAG,...$nvidia/nemotron-3-nano-omni/visionNemotron 3 Nano Omni Vision by NVIDIA - advanced AI model for image-to-text. Delivers high-quality results with fast inference, suitable for both creative and production workflows.Pay per use nvidia/nemotron-3-nano-omni/videoNemotron 3 Nano Omni Video by NVIDIA - advanced AI model for video-to-text. Delivers high-quality results with fast inference, suitable for both creative and production workflows.