v1.1.2

Changelog

Improvements:

• Reasoning with o3 Models: Reasoning support added for OpenAI’s o3 models.

• Gemini embedder update: Updated the GeminiEmbedder to use the new Google’s genai SDK. This update introduces a slight change in the interface:

  # Before
  embeddings = GeminiEmbedder("models/text-embedding-004").get_embedding(
      "The quick brown fox jumps over the lazy dog."
  )
  
  # After
  embeddings = GeminiEmbedder("text-embedding-004").get_embedding(
      "The quick brown fox jumps over the lazy dog."
  )

Bug Fixes:

• Singlestore Fix: Fixed issue where when querying singlestore the embeddings column was returning in binary format.
• MongoDB Vectorstore Fix: MongoDB had multiple issues like creating and dropping the collections twice while initialising. All known issues were fixed.
• LanceDB Fix: Fixed various errors on LanceDB and added on_bad_vectors as parameters.

What's Changed

• Fix singlestore database by @manthanguptaa in #2108
• Fix mongodb vector database by @manthanguptaa in #2125
• cookbook/hackathon by @ysolanky in #2118
• Reasoning using o-series models by @ashpreetbedi in #2117
• hackathon-update-yash by @ysolanky in #2132
• Tic Tac Toe example by @anuragts in #2066
• Podcast Generator by @Itz-Antaripa in #2094
• hackathon-update-examples by @ysolanky in #2138
• Update lancedb and gemini embedder by @dirkbrnd in #2102
• async_delay by @ysolanky in #2141
• Release 1.1.2 by @dirkbrnd in #2140

Full Changelog: v1.1.1...v1.1.2