CHANGELOG.md

CHANGELOG

Key

• [BREAKING]: A breaking change. After an upgrade, your app may need modifications to keep working correctly.
• [FEATURE]: A non-breaking improvement to the app. Either introduces new functionality, or improves on an existing feature.
• [BUGFIX]: Fixes a bug with a non-breaking change.
• [COMPAT]: Compatibility improvements - changes to make Langchain.rb more compatible with different dependency versions.
• [OPTIM]: Optimization or performance increase.
• [DOCS]: Documentation changes. No changes to the library's behavior.
• [SECURITY]: A change which fixes a security vulnerability.

[Unreleased]

• [BUGFIX] [#1044] Cap automatic assistant turns to prevent infinite tool execution loops.
• [BUGFIX] [#1040] Fix parallel_tool_calls: false for Google Gemini assistants by serializing tool execution across model turns.
• [COMPAT] [#980] Suppress a Ruby 3.4 warning for URI parser.
• [BREAKING] [#997] Remove Langchain::Vectorsearch::Epsilla class
• [BREAKING] [#1003] Response classes are now namespaced under Langchain::LLM::Response, converted to Rails engine

[0.19.5]

• [BREAKING] [#859] Add metadata support to PgVector storage
• [BUGFIX] [#939] Fix Langchain::Vectorsearch::Milvus initializer by passing :api_key
• [BUGFIX] [#953] Handle nil response in OpenAI LLM streaming
• [BREAKING] [#956] Deprecate Langchain::Vectorsearch::Epsilla class
• [BREAKING] [#961] Deprecate Langchain::LLM::LlamaCpp class
• [BREAKING] [#962] Deprecate Langchain::LLM::AI21 class
• [BREAKING] [#971] Exclude temperature from being automatically added to OpenAI LLM parameters
• [OPTIM] [#977] Enable Langchain::Tool::RubyCodeInterpreter on Ruby 3.3+

[0.19.4] - 2025-02-17

• [BREAKING] [#894] Tools can now output image_urls, and all tool output must be wrapped by a tool_response() method
• [BUGFIX] [#921] Fix for Assistant when OpenAI o1/o3 models are used

[0.19.3] - 2025-01-13

• [BUGFIX] [#900] Empty text content should not be set when content is nil when using AnthropicMessage

[0.19.2] - 2024-11-26

• [FEATURE] [#884] Add tool_execution_callback to Langchain::Assistant, a callback function (proc, lambda) that is called right before a tool is executed

[0.19.1] - 2024-11-21

• [FEATURE] [#858] Assistant, when using Anthropic, now also accepts image_url in the message.
• [FEATURE] [#861] Clean up passing max_tokens to Anthropic constructor and chat method
• [FEATURE] [#849] Langchain::Assistant now works with AWS Bedrock-hosted Anthropic models
• [OPTIM] [#867] Refactor GoogleGeminiMessage#to_hash and OpenAIMessage#to_hash methods.
• [OPTIM] [#849] Simplify Langchain::LLM::AwsBedrock class
• [BUGFIX] [#869] AnthropicMessage now correctly handles tool calls with content.
• [OPTIM] [#870] Assistant, when using Ollama (e.g.: llava model), now also accepts image_url in the message.

[0.19.0] - 2024-10-23

• [BREAKING] [#840] Rename chat_completion_model_name parameter to chat_model in Langchain::LLM parameters.
• [BREAKING] [#840] Rename completion_model_name parameter to completion_model in Langchain::LLM parameters.
• [BREAKING] [#840] Rename embeddings_model_name parameter to embedding_model in Langchain::LLM parameters.
• [BUGFIX] [#850] Fix MistralAIMessage to handle "Tool" Output
• [BUGFIX] [#837] Fix bug when tool functions with no input variables are used with Langchain::LLM::Anthropic
• [BUGFIX] [#836] Fix bug when assistant.instructions = nil did not remove the system message
• [FEATURE] [#838] Allow setting safety_settings: [] in default_options for Langchain::LLM::GoogleGemini and Langchain::LLM::GoogleVertexAI constructors
• [BUGFIX] [#871] Allow passing in options hash to Ollama

[0.18.0] - 2024-10-12

• [BREAKING] Remove Langchain::Assistant#clear_thread! method
• [BREAKING] Langchain::Messages::* namespace had migrated to Langchain::Assistant::Messages::*
• [BREAKING] Modify Langchain::LLM::AwsBedrock constructor to pass model options via default_options: {...}
• Introduce Langchain::Assistant#parallel_tool_calls options whether to allow the LLM to make multiple parallel tool calls. Default: true
• Minor improvements to the Langchain::Assistant class
• Added support for streaming with Anthropic
• Bump anthropic gem
• Default Langchain::LLM::Anthropic chat model is "claude-3-5-sonnet-20240620" now

[0.17.1] - 2024-10-07

• Move Langchain::Assistant::LLM::Adapter-related classes to separate files
• Fix Langchain::Tool::Database#describe_table method

[0.17.0] - 2024-10-02

• [BREAKING] Langchain::Vectorsearch::Milvus was rewritten to work with newer milvus 0.10.0 gem
• [BREAKING] Removing Langchain::LLM::GooglePalm
• Assistant can now process image_urls in the messages (currently only for OpenAI and Mistral AI)
• Vectorsearch providers utilize the global Langchain.logger
• Update required milvus, qdrant and weaviate versions

[0.16.1] - 2024-09-30

• Deprecate Langchain::LLM::GooglePalm
• Allow setting response_object: {} parameter when initializing supported Langchain::LLM::* classes
• Simplify and consolidate logging for some of the LLM providers (namely OpenAI and Google). Now most of the HTTP requests are being logged when on DEBUG level
• Improve doc on how to set up a custom logger with a custom destination

[0.16.0] - 2024-09-19

• Remove Langchain::Thread class as it was not needed.
• Support cohere provider for Langchain::LLM::AwsBedrock#embed

[0.15.6] - 2024-09-16

• Throw an error when Langchain::Assistant#add_message_callback is not a callable proc.
• Resetting instructions on Langchain::Assistant with Google Gemini no longer throws an error.
• Add Meta models support for AWS Bedrock LLM

[0.15.5] - 2024-09-10 🇧🇦

• Fix for Langchain::Prompt::PromptTemplate supporting nested JSON data
• Require common libs at top-level
• Add add_message_callback to Langchain::Assistant constructor to invoke an optional function when any message is added to the conversation
• Adding Assistant syntactic sugar with #run! and #add_message_and_run!

[0.15.4] - 2024-08-30

• Improve the Langchain::Tool::Database tool
• Allow explictly setting tool_choice on the Assistant instance
• Add support for bulk embedding in Ollama
• Langchain::Assistant works with Langchain::LLM::MistralAI llm
• Fix Langchain::LLM::Azure not applying full default_options

[0.15.3] - 2024-08-27

• Fix OpenAI#embed when text-embedding-ada-002 is used

[0.15.2] - 2024-08-23

• Add Assistant#add_messages() method and fix Assistant#messages= method

[0.15.1] - 2024-08-19

• Drop to_bool gem in favour of custom util class
• Drop colorize which is GPL-licensed in favour of rainbow
• Improve Langchain::Tool::Weather tool

[0.15.0] - 2024-08-14

• Fix Langchain::Assistant when llm is Anthropic
• Fix GoogleGemini#chat method
• Langchain::LLM::Weaviate initializer does not require api_key anymore
• [BREAKING] Langchain::LLM::OpenAI#chat() uses gpt-4o-mini by default instead of gpt-3.5-turbo previously.
• [BREAKING] Assistant works with a number of open-source models via Ollama.
• [BREAKING] Introduce new Langchain::ToolDefinition module to define tools. This replaces the previous reliance on subclassing from Langchain::Tool::Base.

[0.14.0] - 2024-07-12

• Removed TokenLength validators
• Assistant works with a Mistral LLM now
• Assistant keeps track of tokens used
• Misc fixes and improvements

[0.13.5] - 2024-07-01

• Add Milvus#remove_texts() method
• Langchain::Assistant has a state now
• Misc fixes and improvements

[0.13.4] - 2024-06-16

• Fix Chroma#remove_texts() method
• Fix NewsRetriever Tool returning non UTF-8 characters
• Misc fixes and improvements

[0.13.3] - 2024-06-03

• New 🛠️ Langchain::Tool::Tavily to execute search (better than the GoogleSearch tool)
• Remove activesupport dependency
• Misc fixes and improvements

[0.13.2] - 2024-05-20

• New Langchain::LLM::GoogleGemini#embed() method
• Langchain::Assistant works with Langchain::LLM::Anthropic llm
• New XLS file processor
• Fixes and improvements

[0.13.1] - 2024-05-14

• Better error handling for Langchain::LLM::GoogleVertexAI

[0.13.0] - 2024-05-14

• New 🛠️ Langchain::Tool::NewsRetriever tool to fetch news via newsapi.org
• Langchain::Assistant works with Langchain::LLM::GoogleVertexAI and Langchain::LLM::GoogleGemini llms
• [BREAKING] Introduce new Langchain::Messages::Base abstraction

[0.12.1] - 2024-05-13

• Langchain::LLM::Ollama now uses llama3 by default
• Langchain::LLM::Anthropic#complete() now uses claude-2.1 by default
• Updated with new OpenAI models, including gpt-4o
• New Langchain::LLM::Cohere#chat() method.
• Introducing UnifiedParameters to unify parameters across LLM classes

[0.12.0] - 2024-04-22

• [BREAKING] Rename dimension parameter to dimensions everywhere

[0.11.4] - 2024-04-19

• New Langchain::LLM::AWSBedrock#chat() to wrap Bedrock Claude requests
• New Langchain::LLM::OllamaResponse#total_tokens() method

[0.11.3] - 2024-04-16

• New Langchain::Processors::Pptx to parse .pptx files
• New Langchain::LLM::Anthropic#chat() support
• Misc fixes

[0.11.2]

• New Langchain::Assistant#clear_thread! and Langchain::Assistant#instructions= methods

[0.11.1]

• Langchain::Tool::Vectorsearch that wraps Langchain::Vectorsearch::* classes. This allows the Assistant to call the tool and inject data from vector DBs.

[0.11.0]

• Delete previously deprecated Langchain::Agent::ReActAgent and Langchain::Agent::SQLQueryAgent classes
• New Langchain::Agent::FileSystem tool that can read files, write to files, and list the contents of a directory

[0.10.3]

• Bump dependencies
• Ollama#complete fix
• Misc fixes

[0.10.2]

• New Langchain::LLM::Mistral
• Drop Ruby 3.0 support
• Fixes Zeitwerk::NameError

[0.10.1] - GEM VERSION YANKED

[0.10.0]

• Delete Langchain::Conversation class

[0.9.5]

• Now using OpenAI's "text-embedding-3-small" model to generate embeddings
• Added remove_texts(ids:) method to Qdrant and Chroma
• Add Ruby 3.3 support

[0.9.4]

• New Ollama#summarize() method
• Improved README
• Fixes + specs

[0.9.3]

• Add EML processor
• Tools can support multiple-methods
• Bump gems and bug fixes

[0.9.2]

• Fix vectorsearch#ask methods
• Bump cohere-ruby gem

[0.9.1]

• Add support for new OpenAI models
• Add Ollama#chat method
• Fix and refactor of Langchain::LLM::Ollama, responses can now be streamed.

[0.9.0]

• Introducing new Langchain::Assistant that will be replacing Langchain::Conversation and Langchain::Agents.
• Langchain::Conversation is deprecated.

[0.8.2]

• Introducing new Langchain::Chunker::Markdown chunker (thanks @spikex)
• Fixes

[0.8.1]

• Support for Epsilla vector DB
• Fully functioning Google Vertex AI LLM
• Bug fixes

[0.8.0]

• [BREAKING] Updated llama_cpp.rb to 0.9.4. The model file format used by the underlying llama.cpp library has changed to GGUF. llama.cpp ships with scripts to convert existing files and GGUF format models can be downloaded from HuggingFace.
• Introducing Langchain::LLM::GoogleVertexAI LLM provider

[0.7.5] - 2023-11-13

• Fixes

[0.7.4] - 2023-11-10

• AWS Bedrock is available as an LLM provider. Available models from AI21, Cohere, AWS, and Anthropic.

[0.7.3] - 2023-11-08

• LLM response passes through the context in RAG cases
• Fix gpt-4 token length validation

[0.7.2] - 2023-11-02

• Azure OpenAI LLM support

[0.7.1] - 2023-10-26

• Ragas evals tool to evaluate Retrieval Augmented Generation (RAG) pipelines

[0.7.0] - 2023-10-22

• BREAKING: Moving Rails-specific code to langchainrb_rails gem

[0.6.19] - 2023-10-18

• Elasticsearch vector search support
• Fix lib/langchain/railtie.rb not being loaded with the gem

[0.6.18] - 2023-10-16

• Introduce `Langchain::LLM::Response`` object
• Introduce Langchain::Chunk object
• Add the ask() method to the Langchain::ActiveRecord::Hooks

[0.6.17] - 2023-10-10

• Bump weaviate and chroma-db deps
• Langchain::Chunker::Semantic chunker
• Re-structure Conversations class
• Bug fixes

[0.6.16] - 2023-10-02

• HyDE-style similarity search
• Langchain::Chunker::Sentence chunker
• Bug fixes

[0.6.15] - 2023-09-22

• Bump weaviate-ruby gem version
• Ollama support

[0.6.14] - 2023-09-11

• Add find method to Langchain::Vectorsearch::Qdrant
• Enhance Google search output
• Raise ApiError when OpenAI returns an error
• Update OpenAI complete method to use chat completion api
  • Deprecate legacy completion models. See https://platform.openai.com/docs/deprecations/2023-07-06-gpt-and-embeddings

[0.6.13] - 2023-08-23

• Add k: parameter to all ask() vector search methods
• Bump Faraday to 2.x

[0.6.12] - 2023-08-13

[0.6.11] - 2023-08-08

[0.6.10] - 2023-08-01

• 🗣️ LLMs
  • Introducing Anthropic support

[0.6.9] - 2023-07-29

[0.6.8] - 2023-07-21

[0.6.7] - 2023-07-19

• Support for OpenAI functions
• Streaming vectorsearch ask() responses

[0.6.6] - 2023-07-13

• Langchain::Chunker::RecursiveText
• Fixes

[0.6.5] - 2023-07-06

• 🗣️ LLMs
  • Introducing Llama.cpp support
• Langchain::OutputParsers::OutputFixingParser to wrap a Langchain::OutputParser and handle invalid response

[0.6.4] - 2023-07-01

• Fix Langchain::Vectorsearch::Qdrant#add_texts()
• Introduce ConversationMemory
• Allow loading multiple files from a directory
• Add get_default_schema(), create_default_schema(), destroy_default_schema() missing methods to Langchain::Vectorsearch::* classes

[0.6.3] - 2023-06-25

• Add #destroy_default_schema() to Langchain::Vectorsearch::* classes

[0.6.2] - 2023-06-25

• Qdrant, Chroma, and Pinecone are supported by ActiveRecord hooks

[0.6.1] - 2023-06-24

• Adding support to hook vectorsearch into ActiveRecord models

[0.6.0] - 2023-06-22

• [BREAKING] Rename ChainOfThoughtAgent to ReActAgent
• Implement A21 token validator
• Add Langchain::OutputParsers

[0.5.7] - 2023-06-19

• Developer can modify models used when initiliazing Langchain::LLM::* clients
• Improvements to the SQLQueryAgent and the database tool

[0.5.6] - 2023-06-18

• If used with OpenAI, Langchain::Conversation responses can now be streamed.
• Improved logging
• Langchain::Tool::SerpApi has been renamed to Langchain::Tool::GoogleSearch
• JSON prompt templates have been converted to YAML
• Langchain::Chunker::Text is introduced to provide simple text chunking functionality
• Misc fixes and improvements

[0.5.5] - 2023-06-12

• [BREAKING] Rename Langchain::Chat to Langchain::Conversation
• 🛠️ Tools
  • Introducing Langchain::Tool::Weather, a tool that calls Open Weather API to retrieve the current weather

[0.5.4] - 2023-06-10

• 🔍 Vectorsearch
  • Introducing support for HNSWlib
• Improved and new Langchain::Chat interface that persists chat history in memory

[0.5.3] - 2023-06-09

• 🗣️ LLMs
  • Chat message history support for Langchain::LLM::GooglePalm and Langchain::LLM::OpenAI

[0.5.2] - 2023-06-07

• 🗣️ LLMs
  • Auto-calculate the max_tokens: setting to be passed on to OpenAI

[0.5.1] - 2023-06-06

• 🛠️ Tools
  • Modified Tool usage. Agents now accept Tools instances instead of Tool strings.

[0.5.0] - 2023-06-05

• [BREAKING] LLMs are now passed as objects to Vectorsearch classes instead of llm: :name, llm_api_key: previously
• 📋 Prompts
  • YAML prompt templates are now supported
• 🚚 Loaders
  • Introduce Langchain::Processors::Xlsx to parse .xlsx files

[0.4.2] - 2023-06-03

• 🗣️ LLMs
  • Introducing support for AI21
• Better docs generation
• Refactors

[0.4.1] - 2023-06-02

• Beautiful colored log messages
• 🛠️ Tools
  • Introducing Langchain::Tool::RubyCodeInterpreter, a tool executes sandboxed Ruby code

[0.4.0] - 2023-06-01

• [BREAKING] Everything is namespaced under Langchain:: now
• Pgvector similarity search uses the cosine distance by default now
• OpenAI token length validation using tiktoken_ruby

[0.3.15] - 2023-05-30

• Drop Ruby 2.7 support. It had reached EOD.
• Bump pgvector-ruby to 0.2
• 🚚 Loaders
  • Support for options and block to be passed to CSV processor

[0.3.14] - 2023-05-28

• 🔍 Vectorsearch
  • Not relying on Weaviate modules anymore
  • Adding missing specs for Qdrant and Milvus classes
• 🚚 Loaders
  • Add Langchain::Data result object for data loaders
• 🗣️ LLMs
  • Add summarize() method to the LLMs

[0.3.13] - 2023-05-26

• 🔍 Vectorsearch
  • Pgvector support
• 🚚 Loaders
  • CSV loader
  • JSON loader
  • JSONL loader

[0.3.12] - 2023-05-25

• 🔍 Vectorsearch
  • Introduce namespace support for Pinecone
• 🚚 Loaders
  • Loaders overhaul

[0.3.11] - 2023-05-23

• 🗣️ LLMs
  • Introducing support for Google PaLM (Pathways Language Model)
• Bug fixes and improvements

[0.3.10] - 2023-05-19

• 🗣️ LLMs
  • Introducing support for Replicate.com

[0.3.9] - 2023-05-19

• 🚚 Loaders
  • Introduce Loaders::Docx to parse .docx files

[0.3.8] - 2023-05-19

• 🔍 Vectorsearch

  • Introduce support for Chroma DB

• 🚚 Loaders

  • Bug fix Loaders::Text to only parse .txt files

[0.3.7] - 2023-05-19

• 🚚 Loaders
  • Introduce Loaders::Text to parse .txt files
  • Introduce Loaders::PDF to parse .pdf files

[0.3.6] - 2023-05-17

• 🗣️ LLMs
  • Bump hugging-face gem version

[0.3.5] - 2023-05-16

• Bug fixes

[0.3.4] - 2023-05-16

• 🗣️ LLMs
  • Introducing support for HuggingFace

[0.3.3] - 2023-05-16

• Dependencies are now optionally loaded and required at runtime
• Start using standardrb for linting
• Use the Ruby logger

[0.3.2] - 2023-05-15

• 🤖 Agents
  • Fix Chain of Thought prompt loader

[0.3.1] - 2023-05-12

• 🛠️ Tools
  • Introducing Tool::Wikipedia, a tool that looks up Wikipedia entries

[0.3.0] - 2023-05-12

• 🤖 Agents
  • Introducing Agent::ChainOfThoughtAgent, a semi-autonomous bot that uses Tools to retrieve additional information in order to make best-effort informed replies to user's questions.
• 🛠️ Tools
  • Introducing Tool::Calculator tool that solves mathematical expressions.
  • Introducing Tool::Search tool that executes Google Searches.

[0.2.0] - 2023-05-09

• 📋 Prompt Templating
  • Ability to create prompt templates and save them to JSON files
  • Default Prompt::FewShotPromptTemplate
  • New examples added to examples/

[0.1.4] - 2023-05-02

• Backfilling missing specs

[0.1.3] - 2023-05-01

• Initial release