MiMo v2.5 Pro
Model Information
Display Name: MiMo v2.5 Pro
API Model ID: xiaomi/mimo-v2.5-pro
Category: Text To Text
Description: MiMo v2.5 Pro is Xiaomi's flagship reasoning model, ranked #10 globally among 371 AI models. Built on a sparse Mixture-of-Experts architecture with 1.02 trillion total parameters and 42 billion active per token, it delivers frontier-level performance on coding, mathematics, and complex agentic tasks. **Architecture:** - 1.02T total parameters (sparse MoE), 42B active per token - 70 layers (1 dense + 69 MoE) with 384 routed experts, 8 selected per token - Hybrid attention: 10 full attention + 60 sliding window layers (6:1 ratio, 128-token window) - Multi-Token Prediction (MTP): 3 lightweight modules for ~3x output speed - FP8 (E4M3) mixed precision **Key Features:** - 1M token context window for massive document and codebase processing - Advanced chain-of-thought reasoning with reasoning_content field - Function calling, tool use, and structured JSON outputs - SWE-bench Pro: 57.2%% — top-tier agentic coding performance - ClawEval: 64%% Pass³ using 40-60%% fewer tokens than competing frontier models - OpenAI-compatible API - MIT License (fully open-source) **Best For:** - Complex multi-step reasoning and agentic workflows - Long-horizon coding tasks and large codebase analysis - Long document summarization and QA (up to 1M tokens) - Mathematical and scientific computing - Cost-effective frontier-level intelligence
Context Window: 1,000,000 tokens
Max Output: 32,768 tokens
How to Use This Model
To use MiMo v2.5 Pro via the HInow.ai API, use the model ID: xiaomi/mimo-v2.5-pro
API Request Example (Chat/Text)
POST https://api.hinow.ai/v1/chat/completions
Authorization: Bearer YOUR_API_KEY
Content-Type: application/json
{
"model": "xiaomi/mimo-v2.5-pro",
"messages": [
{"role": "user", "content": "Your message here"}
]
}
Pricing
- input: $0.5655
- output: $1.131
Available Parameters
- temperature: Controls randomness (0-2). Default: 0.7 (Options: 0, 0.3, 0.5, 0.7, 1.0, 1.5, 2.0)
- top_p: Nucleus sampling (0-1). Default: 0.9 (Options: 0.1, 0.5, 0.7, 0.9, 0.95, 1.0)
- max_tokens: Max tokens to generate (1-32768) (Options: 512, 1024, 2048, 4096, 8192, 16384, 32768)
- frequency_penalty: Reduce token repetition (0-2). Default: 0 (Options: 0, 0.3, 0.5, 0.7, 1.0, 1.5, 2.0)
- presence_penalty: Penalize repeated topics (0-2). Default: 0 (Options: 0, 0.3, 0.5, 0.7, 1.0, 1.5, 2.0)
- response_format: Output format (Options: text, json_object)
Quick Reference
To use this model, set: "model": "xiaomi/mimo-v2.5-pro"
Featured: No
Documentation: https://hinow.ai/models/xiaomi/mimo-v2.5-pro
API Endpoint: https://api.hinow.ai/v1
