Stay free if you only need complimentary credits for new users and access to playground and model explorer. Upgrade if you need wan 2.7 image-edit at $0.0255 per generation and seedream v4.5 edits at $0.034 per generation. Most solo builders can start free.
Why it matters: Pay-per-generation model can become expensive at high volume compared to dedicated GPU rentals
Available from: Pay-Per-Generation
Why it matters: Limited transparency on enterprise SLAs, uptime guarantees, or rate limits from the public homepage
Available from: Pay-Per-Generation
Why it matters: No bundled subscription tiers shown on the landing page â users must estimate spend from per-call pricing
Available from: Pay-Per-Generation
Why it matters: Quality and capability vary significantly across the model catalog, requiring users to benchmark for their specific use case
Available from: Pay-Per-Generation
Why it matters: Reliance on third-party model providers means features and availability can change when upstream vendors update or deprecate models
Available from: Pay-Per-Generation
Why it matters: Advanced feature not available in free plan.
Available from: Pay-Per-Generation
WaveSpeedAI uses pay-per-generation pricing rather than fixed subscriptions. Featured pricing includes Google image edits at $0.119 per generation (15% off $0.14), additional Google editing models at $0.0595, ByteDance Seedream v4.5 edits at $0.034, Alibaba Wan 2.7 image-edit at $0.0255, and Wan 2.7 image-to-video or video-edit at $0.425 each. A 15% promotional discount is currently applied across the featured model lineup, and a free tier provides complimentary credits for new users to explore models before committing to paid usage.
The platform hosts a large and growing catalog of models, including Google's image editing models, ByteDance's Seedream v4.5 (with standard and sequential editing variants), and Alibaba's Wan 2.7 family covering image-to-video, image-edit, image-edit-pro, video-edit, and text-to-video. The catalog spans text-to-image, image-to-image, image-to-video, video-to-video, and audio generation modalities, and new frontier models are added on an ongoing basis.
Yes â WaveSpeedAI is explicitly positioned as a platform for building AI features, creative tools, and workflows, not just a consumer playground. It exposes a documented API, a model explorer for discovering capabilities, and a unified billing layer so developers can swap between models without integrating each vendor SDK separately. The platform's marketing emphasizes speed and scale, making it appropriate for production workloads where inference latency matters.
Compared to the other model-aggregator platforms in our directory of 870+ AI tools, WaveSpeedAI's model catalog is competitive with Replicate and Fal.ai, and it stands out for hosting Chinese-origin frontier models like ByteDance Seedream and Alibaba Wan that are not always available on Western aggregators. Pricing is transparent on a per-call basis and includes promotional discounts, whereas Replicate often bills by GPU-second. Choose WaveSpeedAI if you specifically need access to the Wan or Seedream model families or want a flat per-generation price.
Yes. The featured catalog includes Alibaba Wan 2.7 image-to-video at $0.425 per generation, Wan 2.7 video-edit at $0.425, and Wan 2.7 text-to-video, alongside numerous other video models in the broader catalog. Video generation is treated as a first-class modality on the platform, and the same unified API pattern used for image models applies, making it straightforward to combine image and video pipelines in a single workflow.
Start with the free plan â upgrade when you need more.
Get Started Free âStill not sure? Read our full verdict â
Last verified March 2026