Benchmark AI Starter Kit getting 429 rate limit errors more frequently than it should

SambaNova Devs Tutorials & Starter Kits
1

When reviewing the results of the benchmark run you see gaps in your graph and a review of your output json reports:

"error_code": "Error while running LLM API calls. Check your model name, LLM API type, env variables and endpoint status",
        "error_msg": "429 Client Error: Too Many Requests for url: https://api.sambanova.ai/v1/chat/completions",
        "start_time": "07:42:31.495616"
2

If you are in this situation please review if you received your API key as part of the “Fast API” program. Those keys are being deprecated so please go into the API section of the SambaNova Cloud and regenerate your API key. Then test your application with the new key.

4

I did regenerate API keys…still it fails with error. if you please fix at least knowledge database assistant.

5

@riteshsaurabh I will open a Jira to the Starter Kit team . Can you provide your errorstack and the frequency of the rate limit errors? As an FYI the free tier rate limit on 405b is now 7 a minute. @james.valentine FYI

-Coby