OpenAI and Anthropic API spending limits: how to set them and what they don't cover

Last updated 19 Oct 2026

Short answer: Both providers let you set a monthly spending limit. On OpenAI, set a monthly spend limit for the organization or a project and switch on Enforce a hard limit; without that switch, requests aren't stopped when the limit is reached [O1]. On Anthropic, set your own limit below your tier's monthly cap on the Billing page in the Claude Console, and give each workspace its own limit [A1]. Both limits are monthly, so a runaway job can still use a large share of the month before it's stopped.

Where CostLatch fits

Provider limits are monthly. For tighter limits per run or per day, you can send requests through a proxy that checks a budget before forwarding each one. That's what CostLatch does (disclosure: CostLatch is my product, and this is its site):

• Per-run, daily and monthly dollar caps for each CostLatch key†, checked before a request is forwarded to OpenAI or Anthropic. Over-cap requests are refused.
• It stops exact repeated identical prompts. It doesn't catch every kind of loop.
• It works with any app or agent that lets you set a custom base URL for the OpenAI Chat Completions API or the Anthropic Messages API. It doesn't support the OpenAI Responses API, embeddings or GET /v1/models, and it doesn't work with Cursor or Claude Code.
• What it can't do: it only covers traffic sent through it. A stolen provider key used directly at OpenAI or Anthropic bypasses any proxy, so keep the provider's own spend limit on.
• Key storage: "Your provider key is encrypted with AES-256-GCM before it's stored, and it's only decrypted in memory to forward your request." "Like any proxy, CostLatch has to decrypt your key to use it, so it isn't zero-knowledge: you're trusting me as the operator and Cloudflare as our host. That's why I recommend a dedicated, revocable key." "Create a dedicated OpenAI project key or Anthropic workspace key just for CostLatch, keep the provider's own spend limit on, and you can revoke it at the provider any time."
• Solo or Pro? Solo: 3 keys, 30 days of usage history. Pro: 20 keys, 90 days.
• Pricing: Solo (US$19/month, 3 keys) and Pro (US$39/month, 20 keys) both start with a 7-day free trial for new customers. A card is required, you won't be charged during the trial, and it then renews monthly until you cancel. Local taxes (e.g. GST/VAT) may be added at checkout.

How do I set a spending limit on the OpenAI API?

• For the whole organization: go to Organization limits, choose Edit spend limit under Spend, enter the monthly spend limit, turn on Enforce a hard limit and save [O1].
• For one project: go to Project settings, then Limits, choose Edit spend limit under Spend, enter the monthly spend limit, turn on Enforce a hard limit and save [O1].
• You need permission to manage that organization or project's settings [O1].
• Spend alerts on their own only send a notification, and API traffic continues. They stay active alongside a hard limit, so you can use them as an early warning [O1].

What happens when an OpenAI limit is reached?

• Affected requests return a 429 error with the code organization_spend_limit_exceeded or project_spend_limit_exceeded [O1].
• OpenAI says enforcement is not instantaneous, so recorded spend can slightly exceed the amount you set [O1].
• Traffic resumes when you raise or remove the limit (after the change propagates), or when the limit resets with the next monthly cycle [O1].
• OpenAI also gives each organization its own approved monthly usage limit based on its usage tier. That's separate from the limits you set [O1].

How do I set a spending limit on the Anthropic API?

• Every Start, Build and Scale tier has a monthly spend cap: US$500, US$1,000 and US$200,000 respectively. Custom-tier organizations have no monthly spend cap [A1].
• To set your own, lower limit: in the Claude Console go to the Billing page, then in Spend limits click Adjust limit (or Set limit) and enter a value. It can't be higher than your tier's cap [A1].
• To protect one project from another, set custom spend limits per workspace. You can't set limits on the default workspace [A1].

What happens when an Anthropic limit is reached?

• Your tier's cap: API usage pauses until 00:00 UTC on the first day of the next month, unless you get a higher limit sooner. Requests return HTTP 429 with enforced_spend_limit_reached, and there's no retry-after header, so automatic retries keep failing [A1].
• A limit you set yourself: requests return HTTP 400 with error type invalid_request_error, and the message says when access resumes. Raise or remove the limit to restore access sooner [A1].

What don't provider spending limits cover?

• They're monthly. The pages cited here describe monthly limits, so one bad night can use a big share of the month's budget before the limit trips [O1] [A1].
• They're not instant. OpenAI says recorded spend can slightly exceed the limit [O1].
• Rate limits aren't dollar limits. Rate limits cap how many requests or tokens you can send per minute, not what you spend, so a loop of valid calls can stay under them [A1].
• Alerts don't stop traffic. An OpenAI spend alert notifies you; only a hard limit stops requests [O1].

FAQ

Does OpenAI have a hard spending limit?

Yes. You can set a monthly spend limit for an organization or a project and turn on Enforce a hard limit. Requests then fail with a 429 error once tracked spend reaches it, though OpenAI says enforcement isn't instantaneous [O1].

Do OpenAI spend alerts stop API usage?

No. A spend alert sends a notification and API traffic continues. Only a hard spend limit stops requests [O1].

What happens when I hit my Anthropic spend limit?

If you hit your tier's monthly cap, requests return a 429 error until 00:00 UTC on the first day of the next month, unless you get a higher limit sooner. If you hit a lower limit you set yourself, requests return a 400 error until you raise or remove it or access resumes [A1].

Can I set a daily or per-run spending limit?

The provider pages cited here describe monthly limits. For a daily or per-run dollar limit you need your own code or a proxy that checks a budget before each request, such as CostLatch†. Keep the provider's monthly limit on as well.

Can I limit spending for one project or workspace?

Yes. OpenAI lets you set a spend limit per project [O1]. Anthropic lets you set custom spend limits per workspace, but not on the default workspace [A1].