Choosing the right Mira model

Mira offers a family of four models, each optimized for specific workloads. This guide helps you pick the right model based on your performance, context, and budget requirements.

Model comparison matrix

When to use each model

mira

The general-purpose workhorse for everyday tasks. Ideal for chatbots, question answering, text summarization, translation, and short code generation. The fastest and most cost-effective model in the family.

• Best for — simple Q&A, summarization, translation, short code snippets
• Context — 32K tokens (~50 pages of text)
• Typical latency — < 1 second to first token

mira-pro

A professional-grade model with an expanded context window and built-in thinking mode. Built for enterprise applications that demand large document processing, deep reasoning, and high accuracy.

• Best for — long document analysis, enterprise applications, detailed code review
• Context — 64K tokens (~100 pages of text)
• Advantage — balances capability and cost for professional use

mira-max

The most powerful model with a maximum context window of 128K tokens. Designed for tasks requiring processing of entire codebases, lengthy legal documents, or complex multi-faceted analysis.

• Best for — entire codebases, legal documents, research papers, complex analysis
• Context — 128K tokens (~200 pages of text)
• Output — up to 16K tokens — the longest output of any model

Decision flowchart

Use this text-based flowchart to quickly identify the right model:

1. Is your input > 64K tokens?
   ├─ Yes → mira-max (includes thinking mode)
   └─ No  → go to step 2

2. Is your input > 32K tokens OR do you need deep reasoning / max accuracy?
   ├─ Yes → mira-pro (includes thinking mode)
   └─ No  → mira (best cost-performance ratio)

Task-to-model matrix

Cost vs performance trade-offs

Model costs scale with their capabilities. For most applications, we recommend starting with mira and upgrading only when output quality is insufficient.

Migrating between models

All Mira models share the same API format. To switch between models, simply change the model parameter in your request. Prompts, tools, and system messages remain compatible.

// Simply change the model parameter
const response = await fetch("https://api.vmira.ai/v1/chat/completions", {
  method: "POST",
  headers: {
    "Authorization": "Bearer sk-mira-YOUR_KEY",
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    model: "mira-pro",  // was "mira" — just change this line
    messages: [{ role: "user", content: "Analyze this document..." }],
  }),
});