curl-X POST 'https://pollo.ai/api/platform/v1/generation/minimax/minimax-h3-max/video'\-H'Content-Type: application/json'\-H'x-api-key: YOUR_API_KEY'\-d'{ "input": { "prompt": "A cinematic shot of a golden retriever running through a field of sunflowers at sunset, warm rim light, shallow depth of field", "resolution": "480p", "duration": 5, "aspectRatio": "auto" }}'
{"input":{"prompt":"A cinematic shot of a golden retriever running through a field of sunflowers at sunset, warm rim light, shallow depth of field","resolution":"480p","duration":5,"aspectRatio":"auto"},"webhookUrl":"https://example.com/webhooks/pollo"}
模式
輸入所選端點和模式的欄位。
場地
類型
必需的
描述
prompt
string
是的
Text prompt describing the content to generate.
resolution
string
不
Output resolution of the generated media (e.g. 720p, 1080p, 2K, 4K). Values are case-insensitive.
Action Previsualization: Test whether a single gesture, reveal, or interaction reads clearly before expanding it into a full scene.
Camera Move Studies: Compare a fixed view with a tracking shot for the same action and choose the more readable result.
Sound-Driven Reaction Clips: Build a short moment around a bell, snap, knock, or spoken cue followed by a visible response.
Still-to-Motion Concepts: Animate an approved illustration or photograph around one controlled movement.
Physical Comedy Beats: Explore a setup and payoff that can be understood without a lengthy explanation.
Micro-Demo Storyboards: Visualize a simple interaction, such as opening a folding object, while checking the clarity of each movement.
如何使用 MiniMax H3 Max API
Get an API Key: Create a Pollo API account and generate your API key from the developer dashboard..
Choose the Model: Send requests to the MiniMax H3 Max endpoint at /generation/minimax/minimax-h3-max/video.
Add Your Inputs: Write the action and camera prompt, adding an image for image-guided generation. Select duration, resolution, and aspectRatio using the current schema.
Generate and Retrieve: Submit the task, poll the returned task status, and download the finished video once it succeeds.
MiniMax H3 Max API 的提示詞最佳實踐
在加入氛圍之前，先讓動作容易想像。讓鏡頭只做一件事，並描述應與關鍵時刻同步的聲音。
簡單的提示詞公式
Starting composition + subject + single action + camera behavior + visible endpoint + sound cue + visual treatment
你應該注意什麼
Action Verbs: Prefer concrete directions such as folds, catches, turns, or reaches over instructions like “make it dynamic.”
Camera Purpose: Explain whether the camera follows a subject, reveals an object, or stays still to make movement easier to judge.
Image Compatibility: Choose motion that follows naturally from the posture and framing in the starting image.
Audible Timing: Tie the important sound to an observable action so you can review synchronization directly.
Single-Variable Tests: Change camera direction or action pacing separately to identify which revision improves the clip.
