← ChangeRadar
llama_index: changelog & breaking changes
Auto-tracked from github.com/run-llama/llama_index. Updated Jun 28, 2026.
v0.14.23
2026-06-24
No breaking-change keywords detected in release notes.
v0.14.22
2026-05-14
⚠ Potential breaking changes
- fix: remove stale exports from public **all** lists ([#21133](https://github.com/run-llama/llama_index/pull/21133))
v0.14.21
2026-04-21
⚠ Potential breaking changes
- Fix Breaking Change in Message Block Buffer Resolution ([#21339](https://github.com/run-llama/llama_index/pull/21339))
v0.14.20
2026-04-03
No breaking-change keywords detected in release notes.
v0.14.19
2026-03-25
⚠ Potential breaking changes
- fix(core): remove incorrect per-node delete calls in index helpers ([#21050](https://github.com/run-llama/llama_index/pull/21050))
- fix: llama-cloud managed index and remove llamaparse reader ([#21043](https://github.com/run-llama/llama_index/pull/21043))
v0.14.18
2026-03-16
⚠ Potential breaking changes
- chore: deprecate python 3.9 once and for all ([#20956](https://github.com/run-llama/llama_index/pull/20956))
- chore: deprecate python 3.9 once and for all ([#20956](https://github.com/run-llama/llama_index/pull/20956))
- chore: deprecate python 3.9 once and for all ([#20956](https://github.com/run-llama/llama_index/pull/20956))
- chore: deprecate python 3.9 once and for all ([#20956](https://github.com/run-llama/llama_index/pull/20956))
- chore: deprecate python 3.9 once and for all ([#20956](https://github.com/run-llama/llama_index/pull/20956))
- chore: deprecate python 3.9 once and for all ([#20956](https://github.com/run-llama/llama_index/pull/20956))
- chore: deprecate python 3.9 once and for all ([#20956](https://github.com/run-llama/llama_index/pull/20956))
- chore: deprecate python 3.9 once and for all ([#20956](https://github.com/run-llama/llama_index/pull/20956))
- chore: deprecate python 3.9 once and for all ([#20956](https://github.com/run-llama/llama_index/pull/20956))
- chore: deprecate python 3.9 once and for all ([#20956](https://github.com/run-llama/llama_index/pull/20956))
v0.14.16
2026-03-10
⚠ Potential breaking changes
- refactor: deprecate asyncio_module in favour of get_asyncio_module ([#20902](https://github.com/run-llama/llama_index/pull/20902))
- Deprecate Preprocess reader: service discontinued ([#20759](https://github.com/run-llama/llama_index/pull/20759))
- fix(lantern,yugabytedb): remove deprecated sessionmaker.close_all() from close() ([#20884](https://github.com/run-llama/llama_index/pull/20884))
- fix(lantern,yugabytedb): remove deprecated sessionmaker.close_all() from close() ([#20884](https://github.com/run-llama/llama_index/pull/20884))
v0.14.15
2026-02-18
⚠ Potential breaking changes
- fix: remove redundant metadata_seperator field from TextNode ([#20649](https://github.com/run-llama/llama_index/pull/20649))
- chore: Remove persistent_connection parameter support, update ([#20714](https://github.com/run-llama/llama_index/pull/20714))
- chore: Remove persistent_connection parameter support, update ([#20714](https://github.com/run-llama/llama_index/pull/20714))
- Fix: remove ORM Collection mix-usage with MilvusClient in Milvus vector store ([#20687](https://github.com/run-llama/llama_index/pull/20687))