👨‍🔬Algorithm change log

Here is a high-level list of all the recent changes we have made to our question answering algorithm and set-up.

12 September 2024

Increase the size of conversational memory 3x to improve responses in longer conversations.

31 May 2024

We found the new GPT-4o model to be more obedient so we adjusted our prompts to reflect this conversatism.

20 May 2024

We moved all users over to use the new GPT-4o model exclusively, at no extra cost.

This means all users have access to OpenAI's most powerful model. It also means that, overall, response times should improve as we no longer need to use GPT-4 turbo as a fallback for GPT-3.5.

31 January 2024

Migrated from GPT-4 to GPT-4 turbo.

We moved to OpenAI's latest, fastest, and most powerful model. The GPT-4 models are currently used as "fallbacks", so if initially, the GPT-3.5 model is unable to answer a query (which is faster), we will re-run the query using the more powerful GPT-4 models.

This also ensures that we are operating on the most up-to-date models when OpenAI deprecates (stops using) their older models in a few months.

You can learn more about the GPT-4 turbo model here.

18 January 2024

Reference "relevancy" improvements made.

This means that when we identify the most relevant pieces of content to answer a question, we then perform a secondary "re-ranking" of those references or sources to determine the most "relevant" sources or references amongst them to better answer your questions.

15 January 2024

Prompt improvements to "follow up" questions, i.e. any question after the 1st question in a conversation.