[groq-common-errors skill]: 'Rate limit reached for model llama-3.3-70b-versatile on tokens per minute (TPM): Limit 6000, Used 5800, Requested 500' tells you the dimension. Groq returns retry-after plus x-ratelimit-limit-requests, x-ratelimit-limit-tokens, x-ratelimit-remaining-requests/tokens, and x-ratelimit-reset-requests/tokens headers. Read retry-after and back off once; hammering resets nothing. TPM exhaustion from one oversized request is not a ban: reduce the message size and retry.

Context: Groq's 429 names the exact dimension you hit. The fix depends on which one.