WaveSpeed added Pruna AI’s P-Video 2 Pro Avatar. It animates a single photo to match a voice track, with prompts controlling expressions, gestures and movements.
It produces up to 3 minutes of speech-driven video, starting at $0.02 per second.
AI media generation platform with 1,000+ top-tier models for fast image, video, and audio generation at scale.
WaveSpeedAI is an AI media generation platform that brings 1,000+ top-tier models together in one place for image, video, and audio generation. Built for speed and scale, it lets creators and developers generate visuals and sound quickly through a simple interface, API, or CLI — without hitting the limits of single-model tools. Recent additions include ByteDance's Seed Audio 1.0 for natural speech, soundscapes, and audio ideas from text.
Pricing model: Freemium
Common questions about WaveSpeed
WaveSpeed added Pruna AI’s P-Video 2 Pro Avatar. It animates a single photo to match a voice track, with prompts controlling expressions, gestures and movements.
It produces up to 3 minutes of speech-driven video, starting at $0.02 per second.
ByteDance's Seed Audio 1.0 is now available on WaveSpeedAI for generating speech and soundscapes from text.
You can choose a preset voice, guide generation with reference audio or an image, and adjust speed, volume, pitch, and output format.
Other products to explore and follow.
The people behind WaveSpeed
Founder and CEO
ByteDance's AI-native cloud platform — tap the AI powering TikTok, including Seedance video and Seedream image models, LLMs, and enterprise APIs.
Fast, low-cost AI inference API for image, video, and audio generation — run top models at scale without managing GPUs.
AI creative studio — generate, animate, and edit visual content with cinematic quality from a single browser-based workspace.

AI creative studio for generating images and videos — state-of-the-art generative AI for visual content creation.
Turn Claude into a creative engine — generate professional images and videos with 30+ models directly inside Claude, Cursor, Perplexity, and any MCP-compatible client.
Run AI models via API — image, video, audio, and language models from the open-source community, hosted and scalable.