Standard Compute

All fixes · OpenAI Codex (Business, Enterprise, Edu) · Quota & billing

Codex “You've hit your usage limit. Contact your workspace owner”: Fix

You've hit your usage limit. Contact your workspace owner for more access.

Your monthly Codex limit or your workspace's credit pool is spent, and on ChatGPT Business, Enterprise and Edu only an owner or admin can raise it. Use “Request increase” or “Notify owner” in the banner, or message your admin. Enterprise and Edu per-user limits reset on the 1st of the month (UTC) or with the billing cycle; a Business pool that is out of credits stays blocked until someone adds credits.

What causes it

  • Your per-user monthly limit is reached: admins can set limits per seat type, group or user, and your limit can block you while the pool still has credits.
  • The workspace credit pool is empty or hit its spend cap. Codex CLI then says “Your workspace is out of credits. Ask your workspace owner to refill in order to continue.” or “You hit your spend cap set by the owner of your workspace.”
  • Codex, ChatGPT Work, ChatGPT for Excel and workspace agents draw from the same allowance and credits, so colleagues' use of other tools counts too.
  • On Enterprise and Edu credit billing, a workspace overage limit of 0 blocks credit use once the pool is empty.

How to fix it

  1. Members: choose “Request increase” or “Notify owner” in the limit banner. Requests land under Usage limits → Pending requests in the admin console and admins get an email. You cannot raise your own limit.
  2. Business owners: at chatgpt.com/admin/billing add credits, turn on automatic reload, or raise the seat-type limit or your per-user override.
  3. Enterprise and Edu admins: at admin.openai.com/usage-limits raise the workspace, group or user limit (temporarily or permanently) or approve the pending request; raise the overage limit or budget under billing.
  4. To keep working now: run Codex CLI with an API key or another OpenAI-compatible provider, which bills outside the workspace allowance.

Running an agent? Parallel subagents and cloud tasks spend a per-user monthly limit several times faster than one chat, so the heaviest Codex users on a team hit it first.

Sources: OpenAI Help: credits and spend controls in Business; OpenAI Help: usage limits in Enterprise and Edu; OpenAI docs: Enterprise usage limits; openai/codex: limit messages (error.rs)

Keep working on a fixed monthly plan

For teams that would rather not ration Codex seats through admin requests, Codex CLI runs on Standard Compute with standardcompute launch codex, or a [model_providers] entry in ~/.codex/config.toml with base_url = "https://api.stdcmpt.com/v1", wire_api = "responses" and model standardcompute. Company plans give each member their own key under one bill. Plans cost a fixed $19 to $249 a month for a set compute budget, with no 5-hour or weekly windows. Each key gets 300 to 2,000 requests a minute by plan, and at 100% of the budget new requests return HTTP 402 until renewal, with no overage charge.

Get your API key How it connects

Plans from $19 a month with $20 of compute. Eligible new accounts can try $0.25 of compute first, no card. Trial terms

FAQ

When does the workspace usage limit reset?

Enterprise and Edu per-user limits use a monthly period by default, resetting on the 1st in UTC, or follow the billing cycle if the admin chose that. Temporary increases end with the current period. On Business, a pool that is out of credits stays blocked until credits are added.

Can I buy credits myself?

No. On Business only workspace owners can buy credits or turn on automatic reload, and on Enterprise and Edu admins control the limits. As a member you request an increase.

Related errors