# Bug Reports **URL:** https://community.perplexity.ai/c/bug-reports/10.md?page=6 [Latest](https://community.perplexity.ai/latest.md) · [Categories](https://community.perplexity.ai/categories.md) · [Tags](https://community.perplexity.ai/tags.md) **Page:** 7 --- ## [About Perplexity Data](https://community.perplexity.ai/t/about-perplexity-data/57)
**Author:** [@vikvang](https://community.perplexity.ai/u/vikvang)\ **Replies:** 1\ **Last updated:** [May 15, 2025, 12:11am UTC](https://community.perplexity.ai/t/about-perplexity-data/57 "2025-05-15T00:11:31Z")
I am using Perplexity to fetch events by location name and date. But I got inaccurate data. Can you please let me know when i hit same prompt hit on preplexity UI I got accurate data and another side when i pass same pr… --- ## [Api request speed is very slow, is there any solution？](https://community.perplexity.ai/t/api-request-speed-is-very-slow-is-there-any-solution/54)
**Author:** [@vikvang](https://community.perplexity.ai/u/vikvang)\ **Replies:** 2\ **Last updated:** [May 15, 2025, 12:11am UTC](https://community.perplexity.ai/t/api-request-speed-is-very-slow-is-there-any-solution/54 "2025-05-15T00:11:07Z")
I found that the website was also slow the first time, but after each time answered questions quickly ， But I need to call this API, https://api.perplexity.ai/chat/completions, each time is slow, how to solve? --- ## [Stream responses are 5s slow to get the first token](https://community.perplexity.ai/t/stream-responses-are-5s-slow-to-get-the-first-token/53)
**Author:** [@vikvang](https://community.perplexity.ai/u/vikvang)\ **Replies:** 1\ **Last updated:** [May 15, 2025, 12:10am UTC](https://community.perplexity.ai/t/stream-responses-are-5s-slow-to-get-the-first-token/53 "2025-05-15T00:10:58Z")
Hello, for my usecase I need very low latencies with llama-3.1-sonar-huge-128k-online but currently the model is very slow to respond in streaming requests. The 70b would already be enough with speedness but is it not co… --- ## [API not returning multiple responses with top\_k \> 1 and top\_p ≈ 1](https://community.perplexity.ai/t/api-not-returning-multiple-responses-with-top-k-1-and-top-p-1/50)
**Author:** [@vikvang](https://community.perplexity.ai/u/vikvang)\ **Replies:** 1\ **Last updated:** [May 15, 2025, 12:10am UTC](https://community.perplexity.ai/t/api-not-returning-multiple-responses-with-top-k-1-and-top-p-1/50 "2025-05-15T00:10:36Z")
Issue Description The Perplexity AI API is not returning multiple responses when the top\_k parameter is set to a value greater than 1 and the top\_p parameter is close to 1. This behavior is inconsistent with the expecte… --- ## [Does Perplexity train on my API requests even with Pro or enterprise Pro account?](https://community.perplexity.ai/t/does-perplexity-train-on-my-api-requests-even-with-pro-or-enterprise-pro-account/49)
**Author:** [@vikvang](https://community.perplexity.ai/u/vikvang)\ **Replies:** 1\ **Last updated:** [May 15, 2025, 12:10am UTC](https://community.perplexity.ai/t/does-perplexity-train-on-my-api-requests-even-with-pro-or-enterprise-pro-account/49 "2025-05-15T00:10:29Z")
On API FAQ, it clearly states that Perplexity uses my API request contents for training their AI. Is this also the case if I have Pro or enterprise Pro account? Is there any plan/way to prevent Perplexity from using … --- ## [Response format issue](https://community.perplexity.ai/t/response-format-issue/45)
**Author:** [@vikvang](https://community.perplexity.ai/u/vikvang)\ **Replies:** 1\ **Last updated:** [May 15, 2025, 12:10am UTC](https://community.perplexity.ai/t/response-format-issue/45 "2025-05-15T00:10:01Z")
I want to use pplx gen some keywords related to user input but seems the output is not as expected sometimes my system prompt is: Your task is to generate a list of words based on the user's prompt. The words should b… --- ## [Model repeating the last word again and again](https://community.perplexity.ai/t/model-repeating-the-last-word-again-and-again/44)
**Author:** [@vikvang](https://community.perplexity.ai/u/vikvang)\ **Replies:** 1\ **Last updated:** [May 15, 2025, 12:09am UTC](https://community.perplexity.ai/t/model-repeating-the-last-word-again-and-again/44 "2025-05-15T00:09:55Z")
Using model='llama-3.1-sonar-large-128k-online', for a functionality to deliver insights and recommendations about the most updated salaries by region and role, and recent or upcoming labour regulations. That model is 70… --- ## [How to Ask for API Help](https://community.perplexity.ai/t/how-to-ask-for-api-help/27)
**Author:** [@Kesku](https://community.perplexity.ai/u/Kesku)\ **Replies:** 0\ **Last updated:** [May 14, 2025, 9:02pm UTC](https://community.perplexity.ai/t/how-to-ask-for-api-help/27 "2025-05-14T21:02:15Z")
When posting API questions, always include: Endpoint URL Request/response (redact keys) Error messages What you’ve tried so far This ensures fast and accurate help. [Previous page](https://community.perplexity.ai/c/bug-reports/10.md?page=5)