Claude Sonnet 3.7 RPM quota not corresponding to reality (10 RPM instead of 125 RPM)
There is a significant discrepancy between documented and actual quotas for Claude 3.7 Sonnet on Amazon Bedrock.
Our quota for our account states 125 RPM, but we're being limited to 1-2 requests per 10 seconds, triggering rate limit errors: "BedrockException - {"message":"Too many tokens, please wait before trying again."}"
Our setup uses the "us." prefix for regional load balancing as instructed. Additionally, when attempting to debug, AWS utilization dashboard shows "Error: Network Failure," preventing us from figuring out what's wrong.
- Newest
- Most votes
- Most comments
Hello.
Please contact AWS Support from the URL below.
I think this problem is difficult to solve with AWS re:Post.
If you have a quota issue, I think you can contact AWS Support for free.
https://support.console.aws.amazon.com/support/home#/
When making an inquiry, I think there will be no problem if you select the items below.
Relevant content
- asked 19 days ago
- AWS OFFICIALUpdated a year ago
