漫画 / 故事板 - Anime Cyclists at Forest Waterfall

This public page preserves the author's original prompt. Review the preview and any published evaluation, then copy the text or open it as a draft. Choose the model, references and current cost in the creation workspace before generating.

Create a cinematic anime-style wide landscape scene of two young female cyclists standing on a mossy stone path in a dense summer forest, looking toward a tall multi-tier waterfall cascading down dark wet rocks into a shallow pool. The composition is viewed from behind at a low-to-mid camera angle, with the two figures in the left and center foreground and the waterfall filling the right-center background. Include exactly 2 cyclists: the left cyclist is tall with very long flowing {argument name="hair color" default="black"} hair, wearing a fitted white short-sleeve cycling jersey, black cycling shorts, black fingerless gloves, white calf socks, and black cycling shoes, standing with her back mostly to the viewer and her head turned slightly toward her companion; the center cyclist is shorter with short {argument name="second cyclist hair color" default="red"} hair, wearing a red-and-white short-sleeve cycling jersey with a large star motif on the back, black cycling shorts, black gloves, black socks, and black cycling shoes, turned three-quarters back toward the viewer with one hand lifted as if casually talking. Surround them with lush deep-green foliage, tree branches, vines, moss-covered boulders, old stone steps, and a small weathered stone post on the left. Use dramatic natural sunlight streaming from the upper right through the canopy, creating bright golden highlights, cool green shadows, sparkling water, mist, and glossy reflections on wet stone. The mood should feel like a still frame from a cycling travel video, serene and adventurous, with high-detail anime character rendering blended into a photorealistic forest background, rich contrast, vibrant greens, crisp water detail, and a 16:9 cinematic aspect ratio. No bicycles visible, no text, no watermark.

Adjust the model, settings and references before generating.

Prefill first, then confirm generation and the current cost.