Raglite

Feat

- replace post-processing with declarative optimization (112)
- compute optimal sentence boundaries (110)
- migrate from poetry-cookiecutter to substrate (98)
- make llama-cpp-python an optional dependency (97)
- add ability to directly insert Markdown content into the database (96)
- make importing faster (86)

Fix

- fix CLI entrypoint regression (111)
- lazily raise module not found for optional deps (109)
- revert pandoc extra name (106)
- avoid conflicting chunk ids (93)

Fix

- remove unnecessary stop sequence (84)

Fix

- fix Markdown heading boundary probas (81)
- improve (re)insertion speed (80)
- **deps**: exclude litellm versions that break get_model_info (78)
- conditionally enable `LlamaRAMCache` (83)

Feat

- add support for Python 3.12 (69)
- upgrade from xx_sent_ud_sm to SaT (74)
- add streaming tool use to llama-cpp-python (71)
- improve sentence splitting (72)

Fix

- improve output for empty databases (68)

Feat

- add MCP server (67)
- let LLM choose whether to retrieve context (62)

Fix

- support pgvector v0.7.0+ (63)