Assistant api showing token over load in processing small text file

i am trying to put transcribed .txt file of video to api and create timestamp-small topic form of texts. I dont know why but i put only 36kb of text which is caption of 10 min video and trid to create indices through this prompt:

Please read the contents of the file named “original_test_request_123.txt” and identify the main topics. " “For each topic, provide a start timestamp, title, and brief summary in a structured JSON format. " “Follow this exact structure in your output:\n\n” “[\n” " {\n” " “start_time”: “”,\n" " “index”: “”,\n" " “summary”: “”\n" " },\n" " …\n" “]\n\n” "Output should contain only the JSON array with start_time , index , and summary fields for each topic. " “Do not add any additional text, explanations, or comments outside of the JSON array.”

but it easily exceeded 30000tpm and cannot get any reply.