Standard Compute – Flat-Rate LLM Subscription for Coding Agents

# Standard Compute > Standard Compute is an LLM gateway for AI coding agents: one OpenAI- and Anthropic-compatible API that routes each request to a cost-efficient model, paid with a monthly plan that has a fixed compute budget. Three things set it apart, in this order: (1) cost optimization: the router sends each step to the cheapest model that can do it, on providers vetted for price, uptime and speed, with automatic fallback; (2) opt-in Europe or United States compliance rules on a company, a team or a single key; (3) prompts and responses are not stored and never used for training. Pricing. Individual plans: Starter $19, Economy $39, Standard $89, Pro $249. Business: Teams is free, with any plan per key including Pro Plus $499, Growth $999, Scale $2,499; Enterprise is $1,499/month with every Enterprise feature. Plans carry no fee; pay as you go is usage + 5% for individuals, 7.5% on Teams and 10% on Enterprise. Each plan carries a fixed monthly compute budget (Starter $19, Economy $39, Standard $89, Pro $249, Pro Plus $499, Growth $999, Scale $2,499). Requests run normally until the budget is used, then return HTTP 402 (budget_exhausted) until the period renews or the user upgrades; there are no overage charges and no 5-hour or weekly windows. For individuals, Pay as you go is optional on Economy and above. Teams is free for companies, with region rules per company, team and key, and single Enterprise features from $199/month. Enterprise ($1,499/month) includes Provider restrictions, Audit log, Request audit export, Signed DPA, SSO (SAML), Uptime SLA and 12-month logs. Setup: in any OpenAI-compatible tool set the base URL to https://api.stdcmpt.com/v1 (United States endpoint: https://us.stdcmpt.com/v1), use your API key, and set the model to standardcompute for smart routing, or select up to 5 models in the dashboard. Claude Code uses the Anthropic Messages API: ANTHROPIC_BASE_URL=https://api.stdcmpt.com (no /v1), or run standardcompute launch claude. Eligibl