Getting consistent "HTTP 429 Too Many Requests" for LlamaParse API

[milap-giantanalytics]

Describe the bug
Getting consistent "HTTP 429 Too Many Requests" errors immediately when submitting documents to LlamaParse API, even with only a single document and no concurrent requests. The rate limiting appears to be overly aggressive, preventing normal usage of the service. Errors occur within 3-4 seconds of request submission, suggesting the rate limit is being triggered before any actual processing begins.

Client:

• Python Library

Additional context
Client configuration used:

LlamaParse(
    api_key=api_key,
    result_type="text",
    preprocess_text=True,
    extract_charts=True,
    premium_mode=True,
)

Error pattern observed:

06:30:06 - HTTP 429 Too Many Requests (retry in 4.0s)
06:30:10 - HTTP 429 Too Many Requests (retry in 4.0s)  
06:30:15 - HTTP 429 Too Many Requests (retry in 4.0s)
06:30:19 - HTTP 429 Too Many Requests (retry in 8.0s)

There is nothing in the LlamaCloud dashboard to indicate if I have exceeded the API rate allowance

[logan-markewich]

Do you have a project and/or org id that your API key is under? Would help in tracking down the requests

[milap-giantanalytics]

sure, here's:

• org id - vefe9ac22-5d86-4f23-bf22-15f1eeef292a
• project - Default

[milap-giantanalytics]

@logan-markewich, is there an SLA timeline for such requests? Our entire pipeline is blocked because llama parsing won't work