Generate images and videos with a single AIsa API key. Covers every image and video model AIsa currently exposes, across three different endpoint paths. 4 image models (Gemini, Wan 2.7, Wan 2.7 Pro, Seedream) and 4 Wan video variants (text-to-video and image-to-video).
Use the AIsa Media Gen skill for my agent. Review https://aisa.one/docs/agent-skills/mediagen and the endpoint list on https://aisa.one/skills/media-gen for capabilities, authentication, pricing, and safety guidance. Configure AIsa API access with AISA_API_KEY. Start with one read-only task, summarize which endpoint you would use, and ask for confirmation before any write, posting, or account-changing action.
Generate an image with Gemini: "A cute red panda, cinematic lighting." Generate an image with Wan 2.7 Pro: "Ultra-detailed product shot, studio lighting." Generate an image with Seedream: "Neo-noir detective portrait" (2048x2048). Create a text-to-video with wan2.7-t2v: "Sweeping shot of a neon cyberpunk skyline." Create an image-to-video with wan2.7-i2v from a reference image: "gentle camera push-in." Poll and download a completed video task.
AISA_API_KEY4 image models: gemini-3-pro-image-preview (Google), wan2.7-image, wan2.7-image-pro (Alibaba), seedream-4-5-251128 (ByteDance). 4 video models: wan2.6-t2v, wan2.7-t2v, wan2.6-i2v, wan2.7-i2v. Image generation returns base64 data. The client script automatically parses and saves the file.