From 7e82e4b71b4085990c5d93d6e991cb5891c6c5eb Mon Sep 17 00:00:00 2001 From: Razvan Dinu Date: Thu, 25 May 2023 19:55:26 +0300 Subject: [PATCH] Updated CHANGELOG. --- CHANGELOG.md | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/CHANGELOG.md b/CHANGELOG.md index 81764c694..397943f24 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -6,13 +6,23 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0 ## [Unreleased] +### Added + +- Support to [connect any LLM](./docs/user_guide/configuration-guide.md#supported-llm-models) that implements the BaseLanguageModel interface from LangChain. +- Support for [customizing the prompts](./docs/user_guide/configuration-guide.md#llm-prompts) for specific LLM models. +- Support for [custom initialization](./docs/user_guide/configuration-guide.md#configuration-guide) when loading a configuration through `config.py`. +- Support to extract [user-provided values](./docs/user_guide/advanced/extract-user-provided-values.md) from utterances. + ### Changed +- Improved the logging output for Chat CLI (clear events stream, prompts, completion, timing information). +- Updated system actions to use temperature 0 where it makes sense, e.g., canonical form generation, next step generation, fact checking, etc. - Excluded the default system flows from the "next step generation" prompt. - Updated langchain to 0.0.167. ### Fixed +- Fixed initialization of LangChain tools. - Fixed the overriding of general instructions [#7](https://github.com/NVIDIA/NeMo-Guardrails/issues/7). - Fixed action parameters inspection bug [#2](https://github.com/NVIDIA/NeMo-Guardrails/issues/2). - Fixed bug related to multi-turn flows [#13](https://github.com/NVIDIA/NeMo-Guardrails/issues/13).