> For the complete documentation index, see [llms.txt](https://changelog.graphlit.dev/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://changelog.graphlit.dev/september-2024/september-3-support-for-web-search-feeds-model-deprecations.md).

# September 3: Support for web search feeds, model deprecations

### New Features

* :bulb: Graphlit now supports web search feeds, using the [Tavily](https://tavily.com/) and [Exa.AI](https://exa.ai/) web search APIs. You can choose the `SEARCH` feed type, and assign your search `text` property, and we will ingest the referenced web pages from the search results.  Optionally, you can select the search service via the `serviceType` property under `search` feed properties.  By default, Graphlit will use the Tavily API.
* :zap: We have deprecated these OpenAI models, according to the future support OpenAI is providing to these legacy models: `GPT35_TURBO`, `GPT35_TURBO_0613`, `GPT35_TURBO_16K`, `GPT35_TURBO_16K_0125`, `GPT35_TURBO_16K_0613`, `GPT35_TURBO_16K_1106`, `GPT4`, `GPT4_0613`, `GPT4_32K`, `GPT4_32K_0613`, `GPT4_TURBO_VISION_128K`, and `GPT4_TURBO_VISION_128K_1106`.  We suggest using GPT-4o or GPT-4o Mini instead.

### Bugs Fixed

* GPLA-2523: Can't ingest from same feed URI multiple times and wait on isFeedDone