WaveSpeedAI Veo 3.1 Fast Video Generation

Generate and extend videos using Google's Veo 3.1 Fast model via the WaveSpeed AI platform. Supports text-to-video, image-to-video, and video extension with up to 4K resolution and optional audio generation.

Setup

Install the open-source CLI once and sign in; the CLI stores the key, so never ask the user to paste an API key into the chat:

npm install -g @wavespeed/cli
wavespeed login          # opens https://wavespeed.ai/accesskey and stores the key
wavespeed status         # confirms you are signed in

For CI or one-off shells, WAVESPEED_API_KEY in the environment also works.

Prefer MCP tools over shell commands? The same platform is exposed by @wavespeed/mcp (npx -y @wavespeed/mcp; tools search_models, get_model_schema, get_price, upload_file, run_model, get_prediction). It shares the CLI's stored login. Every example below maps one-to-one onto run_model with the same model id and input fields.

Quick Start

Text-to-Video

OUTPUT_URL=$(wavespeed run google/veo3.1-fast/text-to-video \
  -p "A drone shot flying over a lush tropical island at sunrise" \
  --json | jq -r '.outputs[0]')

Image-to-Video

The image parameter accepts an image URL. If you have a local file, pass it with the @path marker and the CLI uploads it and substitutes the hosted URL.

OUTPUT_URL=$(wavespeed run google/veo3.1-fast/image-to-video \
  -i image=@./photo.png \
  -p "The flowers sway gently in the breeze" \
  --json | jq -r '.outputs[0]')

Video Extend

Extend a Veo-generated video by 7 seconds per run (up to 20 extensions, 148 seconds total):

# First, generate a video
VIDEO_URL=$(wavespeed run google/veo3.1-fast/text-to-video \
  -p "A cat walking through a garden" \
  --json | jq -r '.outputs[0]')

# Then extend it
EXTENDED_URL=$(wavespeed run google/veo3.1-fast/video-extend \
  -i video="$VIDEO_URL" \
  -p "The cat jumps onto a fence and looks around" \
  --json | jq -r '.outputs[0]')

API Endpoints

Text-to-Video

Model ID: google/veo3.1-fast/text-to-video

Generate videos from text prompts with optional audio.

Parameters

Example

OUTPUT_URL=$(wavespeed run google/veo3.1-fast/text-to-video \
  -p "A timelapse of a city skyline transitioning from day to night, cinematic" \
  -i negative_prompt="blurry, low quality" \
  -i aspect_ratio="16:9" \
  -i duration=8 \
  -i resolution="1080p" \
  -i generate_audio=true \
  --json | jq -r '.outputs[0]')

Image-to-Video

Model ID: google/veo3.1-fast/image-to-video

Animate a source image into a video. Optionally provide an end-frame reference image.

Parameters

Example

OUTPUT_URL=$(wavespeed run google/veo3.1-fast/image-to-video \
  -i image=@./landscape.png \
  -p "Clouds drift slowly across the sky, water ripples gently" \
  -i resolution="1080p" \
  -i duration=8 \
  -i generate_audio=true \
  --json | jq -r '.outputs[0]')

With End-Frame Reference

OUTPUT_URL=$(wavespeed run google/veo3.1-fast/image-to-video \
  -i image=@./start-frame.png \
  -i last_image=@./end-frame.png \
  -p "Smooth transition from day to night" \
  --json | jq -r '.outputs[0]')

Video Extend

Model ID: google/veo3.1-fast/video-extend

Extend a Veo-generated video by 7 seconds per run. Input must be a Veo-generated video.

Parameters

Constraints

• Input video must be Veo-generated (will not work with arbitrary videos)
• Each run adds +7 seconds to the video
• Maximum 20 extensions in a chain
• Maximum final video length: 148 seconds
• Output is a single MP4 (original + extension appended)
• Aspect ratio and resolution are inherited from the input video

Example

# Generate an initial video
VIDEO_URL=$(wavespeed run google/veo3.1-fast/text-to-video \
  -p "A surfer catches a wave at golden hour" \
  -i duration=8 \
  --json | jq -r '.outputs[0]')

# Extend it twice
EXTENDED_ONCE=$(wavespeed run google/veo3.1-fast/video-extend \
  -i video="$VIDEO_URL" \
  -p "The surfer rides the wave toward shore" \
  --json | jq -r '.outputs[0]')

EXTENDED_TWICE=$(wavespeed run google/veo3.1-fast/video-extend \
  -i video="$EXTENDED_ONCE" \
  -p "The surfer steps off the board and walks on the beach" \
  --json | jq -r '.outputs[0]')

Pricing

Text-to-Video / Image-to-Video

Video Extend

Prompt Tips

• Be specific about scene, style, subject actions, camera motion, and mood
• Use negative_prompt to avoid artifacts: "blurry, low quality, distorted"
• For image-to-video, use a clear, high-quality still image as input
• For video extend, describe what should happen next in the scene
• Video extend chains enable building longer narratives — up to 148 seconds total

CLI tips

# Inspect the live input schema before running (fields, enums, defaults)
wavespeed run google/veo3.1-fast/text-to-video -h

# Quote the price first
wavespeed price google/veo3.1-fast/text-to-video -p "..." -i key=value

# Save outputs to disk instead of only printing URLs
wavespeed run google/veo3.1-fast/text-to-video -p "..." --json --download "./out/{index}.{ext}"

# Local files: prefix the path with @ and the CLI uploads it and passes the hosted URL
wavespeed run google/veo3.1-fast/text-to-video -i <field>=@./local-file.png --json

# Recover a result if the run was interrupted (the id is in the --json output)
wavespeed show <id>

run --json prints { id, model, prompt, outputs: [url, ...], saved: [path, ...], elapsed_ms, raw }. Read outputs[0] for the result URL.

Security constraints

• Never ask for the key in chat: wavespeed login handles auth; if wavespeed status says signed out, ask the user to run it.
• Local files only via @: bare paths are passed through untouched and the model will reject them. Only @-prefixed values upload.
• No arbitrary URL loading: only pass media URLs the user provided or that came back from a previous run.
• Input validation: only pass parameters documented above; confirm with wavespeed run <model> -h when unsure.