claude code not working on Mac Os with Bedrock

Hi @massi-ang , can you share the command/envvars you used to invoke claude ?

Update:

I tested the tool again today, and it is working as expected, at least for /init.

Same for me on linux. Started with: CLAUDE_CODE_USE_BEDROCK=1 ANTHROPIC_MODEL='us.anthropic.claude-3-7-sonnet-20250219-v1:0' DISABLE_PROMPT_CACHING=1 claude

Keep getting "Too many tokens, please wait before trying again." and then sometime it works.

We suggest getting at least 300k ITPM (input tokens per minute) of Sonnet 3.7 for a good experience. Do you know what your current rate limits are?

Same issue for me on WSL with limit of 1M tokens

@nalindley can you take a screenshot of your rate limits and provide them here?

Same issue :

⎿  API Error (429 Too many tokens, please wait before trying again.) · Retrying in 1 seconds… (attempt 1/10)
  ⎿  API Error (429 Too many tokens, please wait before trying again.) · Retrying in 1 seconds… (attempt 2/10)
  ⎿  API Error (429 Too many tokens, please wait before trying again.) · Retrying in 2 seconds… (attempt 3/10)
  ⎿  API Error (429 Too many tokens, please wait before trying again.) · Retrying in 1 seconds… (attempt 1/10)
  ⎿  API Error (429 Too many tokens, please wait before trying again.) · Retrying in 1 seconds… (attempt 2/10)
  ⎿  API Error (429 Too many tokens, please wait before trying again.) · Retrying in 2 seconds… (attempt 3/10)
  ⎿  API Error (429 Too many tokens, please wait before trying again.) · Retrying in 4 seconds… (attempt 4/10)
  ⎿  API Error (429 Too many tokens, please wait before trying again.) · Retrying in 8 seconds… (attempt 5/10)
  ⎿  API Error (429 Too many tokens, please wait before trying again.) · Retrying in 16 seconds… (attempt 6/10)
  ⎿  A

It was s small task and I already get this issue

Managed to figure it out - talked to AWS support, it's just a capacity issue on their end and they're applying limits lower than the stated service quota unless you have provisioned throughput. They recommended switching to a different region temporarily and that's worked so far for me.

I'm using oregon and which region did you switch to ?

Actually i switched to Oregon - the other thing that might help is just to make sure you've requested access to the model across multiple regions, so switch to us-east-1 and request it, same for us-east-2, etc. I might be making this up, but I think the cross-region inference isn't actually enabled until you have access everywhere

Same issue

Facing same issue, apparently I have too low ITPM (10K only) configured in Service Quotas for Sonnet 3.7, anybody got that increased?

Facing same issue, apparently I have too low ITPM (10K only) configured in Service Quotas for Sonnet 3.7, anybody got that increased?

We can't.

This quota isn't adjustable and you can't request a quota increase.

Sorry I suggest contacting AWS support for increased quota. Or try our API directly at https://console.anthropic.com/

I contacted Support Center and they mentioned currently this limit (10k) cannot be increased.

Do you think downgrading the model version (Claude 3.5) help ?

This issue has been automatically locked since it was closed and has not had any activity for 7 days. If you're experiencing a similar issue, please file a new issue and reference this one if it's relevant.