# MiniMax models: APIs and prices

MiniMax text models with 1M context, plus Hailuo video. XiaoZhan AIhub offers 8 MiniMax models (Text generation, Video generation). Call them with one API key through the OpenAI, Anthropic or Gemini SDK; prices are in credits (60 credits = ¥1) and every model page lists the full price.

- Page: https://qd.zsopc.com/en/vendors/minimax
- Models: 8
- Last updated: 2026-11-17

## Text generation

### MiniMax M2.7

MiniMax M2.7. 200K context.

- Model ID: `MiniMax-M2.7`
- Page: https://qd.zsopc.com/en/models/MiniMax-M2.7 (Markdown: https://qd.zsopc.com/en/models/MiniMax-M2.7.md)
- Context window: 200,000 tokens
- Updated: 2026-11-17
- Price:
  - Input: 121.32 credits (≈ ¥2.02 / ≈ $0.3) per 1M tokens
  - Output: 606.6 credits (≈ ¥10.11 / ≈ $1.5) per 1M tokens
  - Cached input: 24.26 credits (≈ ¥0.404 / ≈ $0.06) per 1M tokens

## Video generation

### Hailuo H3

MiniMax Hailuo H3 video generation (asynchronous task) with native stereo audio. 4-15 seconds, 768P / 2K, aspect ratios 21:9 / 16:9 / 4:3 / 1:1 / 3:4 / 9:16 / adaptive. Text-to-video, first-frame and first-and-last-frame image-to-video, and all-round reference (up to 9 reference images, 3 reference videos and 3 reference audio clips).

- Model ID: `minimax-h3`
- Page: https://qd.zsopc.com/en/models/minimax-h3 (Markdown: https://qd.zsopc.com/en/models/minimax-h3.md)
- Updated: 2026-11-17
- Price:
  - Per reference image: 12 credits (≈ ¥0.2 / ≈ $0.03)
  - 768p, per second of output: 30 credits (≈ ¥0.5 / ≈ $0.074)
  - 2k, per second of output: 48 credits (≈ ¥0.8 / ≈ $0.119)

### Hailuo H3 Fast with IR tx

MiniMax Hailuo H3 video generation (asynchronous task); native audio can be turned off. 4-15 seconds, 768P / 720P / 480P, aspect ratios 21:9 / 16:9 / 4:3 / 1:1 / 3:4 / 9:16 / adaptive. Text-to-video, first-frame, last-frame and first-and-last-frame image-to-video, and all-round reference (up to 9 reference images, 3 reference videos and 3 reference audio clips).

- Model ID: `minimax-h3-fast-ir-tx`
- Page: https://qd.zsopc.com/en/models/minimax-h3-fast-ir-tx (Markdown: https://qd.zsopc.com/en/models/minimax-h3-fast-ir-tx.md)
- Updated: 2026-11-17
- Price:
  - Per reference image: 12 credits (≈ ¥0.2 / ≈ $0.03)
  - 480p, per second of output: 15.6 credits (≈ ¥0.26 / ≈ $0.039)
  - 768p, per second of output: 24 credits (≈ ¥0.4 / ≈ $0.059)
  - 720p, per second of output: 24 credits (≈ ¥0.4 / ≈ $0.059)

### Hailuo H3 Fast tx

MiniMax Hailuo H3 video generation (asynchronous task); native audio can be turned off. 4-15 seconds, 768P / 720P / 480P, aspect ratios 21:9 / 16:9 / 4:3 / 1:1 / 3:4 / 9:16 / adaptive. Text-to-video, first-frame, last-frame and first-and-last-frame image-to-video, and all-round reference (up to 9 reference images, 3 reference videos and 3 reference audio clips).

- Model ID: `minimax-h3-fast-tx`
- Page: https://qd.zsopc.com/en/models/minimax-h3-fast-tx (Markdown: https://qd.zsopc.com/en/models/minimax-h3-fast-tx.md)
- Updated: 2026-11-17
- Price:
  - Per reference image: 12 credits (≈ ¥0.2 / ≈ $0.03)
  - 480p, per second of output: 15.6 credits (≈ ¥0.26 / ≈ $0.039)
  - 720p, per second of output: 24 credits (≈ ¥0.4 / ≈ $0.059)
  - 768p, per second of output: 24 credits (≈ ¥0.4 / ≈ $0.059)

### Hailuo H3 Max

MiniMax Hailuo H3 Max, express video generation (asynchronous task) with native stereo audio. 5-15 seconds, 480P / 768P, aspect ratios 21:9 / 16:9 / 4:3 / 1:1 / 3:4 / 9:16 / adaptive. Text-to-video and first-frame or first-and-last-frame image-to-video (no all-round reference).

- Model ID: `minimax-h3-max`
- Page: https://qd.zsopc.com/en/models/minimax-h3-max (Markdown: https://qd.zsopc.com/en/models/minimax-h3-max.md)
- Updated: 2026-11-17
- Price:
  - 480p, per second of output: 19.8 credits (≈ ¥0.33 / ≈ $0.049)
  - 768p, per second of output: 30 credits (≈ ¥0.5 / ≈ $0.074)

### Hailuo H3 Full with IR tx

MiniMax Hailuo H3 video generation (asynchronous task); native audio can be turned off. 4-15 seconds, 768P / 720P, aspect ratios 21:9 / 16:9 / 4:3 / 1:1 / 3:4 / 9:16 / adaptive. Text-to-video, first-frame, last-frame and first-and-last-frame image-to-video, and all-round reference (up to 9 reference images, 3 reference videos and 3 reference audio clips).

- Model ID: `minimax-h3-medium-ir-tx`
- Page: https://qd.zsopc.com/en/models/minimax-h3-medium-ir-tx (Markdown: https://qd.zsopc.com/en/models/minimax-h3-medium-ir-tx.md)
- Updated: 2026-11-17
- Price:
  - Per reference image: 12 credits (≈ ¥0.2 / ≈ $0.03)
  - 768p, per second of output: 30 credits (≈ ¥0.5 / ≈ $0.074)
  - 720p, per second of output: 30 credits (≈ ¥0.5 / ≈ $0.074)

### Hailuo H3 Full tx

MiniMax Hailuo H3 video generation (asynchronous task); native audio can be turned off. 4-15 seconds, 768P / 720P, aspect ratios 21:9 / 16:9 / 4:3 / 1:1 / 3:4 / 9:16 / adaptive. Text-to-video, first-frame, last-frame and first-and-last-frame image-to-video, and all-round reference (up to 9 reference images, 3 reference videos and 3 reference audio clips).

- Model ID: `minimax-h3-medium-tx`
- Page: https://qd.zsopc.com/en/models/minimax-h3-medium-tx (Markdown: https://qd.zsopc.com/en/models/minimax-h3-medium-tx.md)
- Updated: 2026-11-17
- Price:
  - Per reference image: 12 credits (≈ ¥0.2 / ≈ $0.03)
  - 720p, per second of output: 30 credits (≈ ¥0.5 / ≈ $0.074)
  - 768p, per second of output: 30 credits (≈ ¥0.5 / ≈ $0.074)

### Hailuo H3 tx

MiniMax Hailuo H3 video generation (asynchronous task); native audio can be turned off. 4-15 seconds, 768P / 720P, aspect ratios 21:9 / 16:9 / 4:3 / 1:1 / 3:4 / 9:16 / adaptive. Text-to-video, first-frame, last-frame and first-and-last-frame image-to-video, and all-round reference (up to 9 reference images, 3 reference videos and 3 reference audio clips).

- Model ID: `minimax-h3-tx`
- Page: https://qd.zsopc.com/en/models/minimax-h3-tx (Markdown: https://qd.zsopc.com/en/models/minimax-h3-tx.md)
- Updated: 2026-11-17
- Price:
  - Per reference image: 12 credits (≈ ¥0.2 / ≈ $0.03)
  - 720p, per second of output: 30 credits (≈ ¥0.5 / ≈ $0.074)
  - 768p, per second of output: 30 credits (≈ ¥0.5 / ≈ $0.074)

Prices are in credits, 60 credits = ¥1. The prices above are what is charged today (limited-time discounts included); ≈¥ / ≈$ use the platform's published rates.

## FAQ

### How do I call MiniMax models?

Create an API key on XiaoZhan AIhub, point your SDK's base_url at https://qd.zsopc.com/v1 and put the model ID from the list above in the model field. The OpenAI, Anthropic and Gemini SDKs all work.

### How are MiniMax models billed?

Each model is billed its own way: text models per input and output token, image models per image, video models per second or per token. Prices are in credits (60 credits = ¥1); each model page lists the full price.
