DoublewordDoubleword

Model Name

XiaomiMiMo/MiMo-V2.5-Pro

MiMo V2.5 Pro

  • Type: Generation
  • Capabilities: reasoning, enhanced_structured_generation
  • Cache read: $0.00 per 1M input tokens (0.01× standard input price). See prompt caching.

Overview

Xiaomi’s flagship 1.02T-parameter MoE reasoning model for agentic workflows, complex software engineering, long-horizon tool use, and long-context tasks.

Pricing

PriorityInput Tokens (per 1M)Cache Read Tokens (per 1M)Output Tokens (per 1M)
Realtime1$0.44$0.00$0.87
Async$0.33$0.00$0.65
Batch (24h)$0.22$0.00$0.44

Playground

Open this model in the Playground.

Footnotes

  1. Realtime availability is limited. Doubleword is primarily a batch API.