Video and audio generation with aggressive per-call pricing.
Short-form video, ad creative and audio-visual generation.
Proprietary hosted API.
No weights are available. You are contracting a hosted service, so data flow and terms sit with the provider.
Licence breakdownPrices are live from the catalogue and may change without notice.
| Model | Billing | Input / 1M | Output / 1M | Cached / 1M | Endpoints |
|---|---|---|---|---|---|
| MiniMax-M2.5 A flagship model combining leading coding, agent tool use, search and office performance with high speed and low cost. | Tokens | $0.3134 | $1.25 | $0.0313 | openai, anthropic |
| MiniMax-M2.7 A next-generation language model with multi-agent collaboration for autonomous complex production tasks and self-improvement. | Tokens | $0.3134 | $1.25 | $0.0627 | openai, anthropic |
| MiniMax-M3 Frontier coding capabilities, a one-million-token context and native multimodality. | Dynamic | $2.1 | $8.4 | $0.42 | openai, anthropic |
| minimax_h3_zm_u24 MiniMax H3 multi-image, multi-audio video (enhanced quality): combine up to 9 reference images and 3 reference audio clips to blend characters, products, scenes and sound into one video with coherent motion, natural camera work and synced audio. Supports 1–15 second clips at 480p or 768p in portrait, landscape or 1:1, billed per second. Ideal for character interactions, product ads, story scenes and short-form video. | Tokens | $75 | $75 | — | openai-video |
| minimax_h3_b99_002 MiniMax H3 first-and-last-frame video: upload a first frame and a last frame (JPG/PNG/WebP) plus a prompt, and the model infers the motion path, character actions, camera changes and scene transition in between, producing a smooth, detailed bridging clip. Supports 1–15 second clips in 736p portrait, landscape or 1:1, billed per second. Ideal for AI animation, ad creative, storyboarding, product demos and transitions. | Tokens | $75 | $75 | — | openai-video |
| minimax_h3_b99_001 MiniMax H3 text-to-video: turn a single prompt into cinematic AI video with no reference images. It understands natural-language descriptions of characters, scenes, motion, camera work and lighting, producing smooth, temporally consistent footage. Supports 1–15 second clips in 736p portrait, landscape or 1:1, billed per second. Ideal for short films, brand marketing, product showcases, social content and previsualization. | Tokens | $75 | $75 | — | openai-video |