v4.5.0

What's Changed

• Delaying of state mutation in Docs.aadd_texts by @jamesbraza in #264

Full Changelog: v4.4.1...v4.5.0

v4.4.1

What's Changed

• Creating Answer.get_unique_docs_from_contexts helper method by @jamesbraza in #261
• Added batching to embed documents by @whitead in #262

Full Changelog: v4.4.0...v4.4.1

v4.4.0

What's Changed

• Documenting LLMResult by @jamesbraza in #255
• Migrating from setup.py to pyproject.toml by @jamesbraza in #252
• Adjust whitespace on detailed citations by @whitead in #257

Full Changelog: v4.3.0...v4.4.0

v4.3.0

What's Changed

• Separate parsing and chunking in read_doc by @mskarlin in #254

New Contributors

• @mskarlin made their first contribution in #254

Full Changelog: v4.2.1...v4.3.0

v4.2.1

What's Changed

• Fixed bug for mixed encoding types from numpy by @whitead in #253

Full Changelog: v4.2.0...v4.2.1

v4.2.0

What's Changed

• Upgrade of ruff, adding codespell, toml-sort by @jamesbraza in #251
• Added convenience methods for building sparse/hybrid vector dbs by @whitead in #250

Full Changelog: v4.1.1...v4.2.0

v4.1.1

What's Changed

• Fixed MMR alg on not correctly choosing most similar

Full Changelog: v4.1.0...v4.1.1

v4.1.0

What's Changed

• delete the texts by @Dustyposa in #181
• Making Doc hashable by @jamesbraza in #244
• Latest black and prettier by @jamesbraza in #247
• Add more vector embedding types by @whitead in #245
• Allowing Doc subclasses to propagate to evidence by @jamesbraza in #246
• Fixing crash when json.loads returns str by @jamesbraza in #248

New Contributors

• @Dustyposa made their first contribution in #181

Full Changelog: v4.0.0...v4.1.0

v4.0.0

What's Changed

• Version 4 with better open source model support and no langchain by @whitead in #223
• Prevent error loading pdf: Added pymupdf/fitz open to get page count by @ekcomputer in #225
• Removed nest_asyncio by @whitead in #227
• JSON Mode & Callbacks by @whitead in #229
• Removed features that required python 3.11 by @whitead in #230
• Added status info on used contexts by @whitead in #231
• Improved json parsing by @whitead in #232
• Fixed deserialized computed fields issue by @whitead in #233
• Propogated extra props in JSON by @whitead in #235
• Added py.typed metadata file by @jamesbraza in #238
• mypy config update by @jamesbraza in #237
• More verbose comment on Answer.token_counts by @jamesbraza in #240
• Added anthropic LLMs by @whitead in #241
• Adjusted anthropic import by @whitead in #242

New Contributors

• @ekcomputer made their first contribution in #225
• @jamesbraza made their first contribution in #238

Full Changelog: v3.13.4...v4.0.0

v4.0.0-pre.11

What's Changed

• Adjusted anthropic import by @whitead in #242

Full Changelog: v4.0.0-pre.10...v4.0.0-pre.11