Wan 2.7 Reference-to-Video

Reference-driven video generation for single-character, multi-character, and multi-shot workflows

About model

On Together AI, Wan 2.7 Reference-to-Video costs $0.10 per video.

Wan 2.7 Reference-to-Video is a generation model that uses image and video references to drive outputs for single-character, multi-character, and multi-shot workflows, with native 1080p outputs from 2 to 10 seconds.

Reference Inputs

Images/Video

Image and video inputs

Duration

2-10s

Supported generation length

Control

Multi-shot

Single and multi-shot workflows

Model key capabilities

Reference Inputs: Supports reference image and video inputs
Workflow Modes: Single-shot and multi-shot reference-driven generation
Mixed Modes: Mixed reference-image and reference-video workflows

API usage

cURL
Python
Typescript

Endpoint:

Wan-AI/wan2.7-r2v

curl --request POST \
  --url https://api.together.ai/v2/videos \
  --header "Authorization: Bearer $TOGETHER_API_KEY" \
  --header "Content-Type: application/json" \
  --data '{
    "model": "Wan-AI/wan2.7-r2v",
    "prompt": "some penguins building a snowman"
  }'

from together import Together

client = Together()

# Create a video generation job
job = client.videos.create(
    prompt="A serene sunset over the ocean with gentle waves",
    model="Wan-AI/wan2.7-r2v"
)

import Together from "together-ai";

const together = new Together();

async function main() {
  // Create a video generation job
  const job = await together.videos.create({
    prompt: "A serene sunset over the ocean with gentle waves",
    model: "Wan-AI/wan2.7-r2v"
  });

Model card
Architecture Overview:
• Reference-to-video (R2V) generation model
• Reference Support: Driven by reference image and video inputs
• Workflow Support: Single-shot and multi-shot output modes
• Native 1080p output

Performance Characteristics:
• Supports 2-10 second generation
‍
Prompting
Together AI API Access:
• Access Wan 2.7 Reference-to-Video via Together AI APIs using the endpoint Wan-AI/wan2.7-r2v
• Authenticate using your Together AI API key in request headers
• Reference inputs accept images and video
• Available on Together AI serverless infrastructure
‍
Applications & use cases
Creative Production:
• Multi-subject scene workflows with reference image and video inputs
• Single-shot and multi-shot reference-driven generation
• Mixed reference-image and reference-video workflows
‍