tavily-ai/tavily-python

Variability in Results

Opened this issue · 1 comments

bookandlover commented

we are experiencing the following issues:

1.	Variability in Results: Each API call returns different results, even when the same query parameters are used. This lack of consistency makes it difficult for us to maintain the reliability of our data.
2.	Quality of Data: The relevance and quality of the data retrieved fluctuate significantly, which affects the integrity of our analyses and operations.

We would greatly appreciate your guidance on how to address these issues. Are there specific configurations or best practices we should follow to ensure more consistent and high-quality results? Additionally, could you provide any insights into potential causes for these variations?

carl-vbn commented

Hello,
We recently deployed an update to the API to address the first issue. Please let us know if you are still experiencing strong variability.
Regarding your second point, we are constantly working on improving our systems to provide the best possible results. We apologize for any negative experiences you may have faced while using the API. You should see an increase in quality as time goes by. We also encourage you to experiment with different prompts. Wording and semantics can make a big difference.