Standard Reason
72B.

Overview

Standard Reason 72B is our highest-capability reasoning model for production workloads. It is tuned to decompose difficult tasks, maintain state across long contexts, and return structured, verifiable answers.

Use it for multi-step research, complex code generation, planning, analysis, and agent orchestration where accuracy matters more than lowest possible latency.

Pricing

Usage is billed by token with no monthly model fee. Cached input receives a lower rate when the same context prefix is reused.

Specifications

Benchmarks below reflect the shared production deployment. Dedicated and private clusters can be provisioned with higher throughput targets.

Supported capabilities

Standard Reason 72B supports the OpenAI-compatible chat completions interface and production features used by modern agents.

• Streaming
• Tool calling
• JSON mode
• Structured outputs
• Reasoning effort
• Long context
• System prompts
• Batch API

API example

Use the existing OpenAI SDK, change the base URL, and select the model ID.

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.STANDARD_API_KEY,
  baseURL: "https://api.standardthinking.ai/v1",
});

const response = await client.chat.completions.create({
  model: "standard-reason-72b",
  messages: [{ role: "user", content: "Explain this architecture." }],
  stream: true,
});