Skip to content

Claude Haiku 4.5 token quota set to 0 and non-adjustable in eu-central-1 — ThrottlingException on first call

-1

I'm getting ThrottlingException: Too many tokens per day on the very first API call of the day to eu.anthropic.claude-haiku-4-5-20251001-v1:0 in eu-central-1. Root cause identified: the quota "Model invocation max tokens per day for Anthropic Claude Haiku 4.5" is set to 0 in my account, and is marked Adjustable = false, so I cannot fix it myself. Model status: AUTHORIZED and AVAILABLE. Inference profile: ACTIVE. IAM: AmazonBedrockFullAccess. All self-service remediation paths exhausted. Support case 177902340800109 has been open and unassigned for 14 days. Requesting urgent attention — we are on a trial account and this is a complete blocker.

1 Answer
0

I understand you're experiencing a critical issue with Claude Haiku 4.5 in eu-central-1 where your token quota is set to 0 and cannot be adjusted, blocking your ability to use the model despite it being authorized and available.

For quota increases and account-specific limitations that cannot be self-adjusted through the AWS Service Quotas console, you'll need AWS Support assistance. Since you mention having an open support case (177902340800109) that has been unassigned for 14 days, I recommend the following escalation steps:

  1. Reply to your existing support case with updated details emphasizing the business impact and that this is a complete blocker for your trial evaluation.

  2. Check your support plan level - Trial accounts typically have Basic Support, which provides limited technical support. Consider temporarily upgrading to Developer or Business Support for faster response times on technical issues like quota adjustments.

  3. Use the AWS Support Center to check if there are any updates on your case or if additional information is needed from your side.

  4. Verify account standing - Ensure there are no billing or verification issues with your trial account that might be causing automatic quota restrictions.

  5. Consider alternative regions temporarily - While not ideal, you might test if other regions have different quota settings to unblock your evaluation while waiting for support resolution.

For urgent production-impacting issues, Business or Enterprise Support plans provide faster response times and the ability to escalate cases. Since this is blocking your trial evaluation, emphasizing the time-sensitive nature in your support case communications is important.

answered 8 days ago
EXPERT
reviewed 8 days ago

You are not logged in. Log in to post an answer.

A good answer clearly answers the question and provides constructive feedback and encourages professional growth in the question asker.