v2.25.0

2.25.0

(Full Changelog)

Enhancements made

• Export context hooks from NPM package entry point #1020 (@dlqqq)
• Add support for optional telemetry plugin #1018 (@dlqqq)
• Add back history and reset subcommand in magics #997 (@akaihola)

Maintenance and upkeep improvements

Contributors to this release

(GitHub contributors page for this release)

@akaihola | @dlqqq | @jtpio | @pre-commit-ci

v2.24.1

2.24.1

(Full Changelog)

Enhancements made

• Make path argument required on /learn #1012 (@andrewfulton9)

Bugs fixed

• Export tokens from lib/index.js #1019 (@dlqqq)

Contributors to this release

(GitHub contributors page for this release)

@andrewfulton9 | @dlqqq | @hockeymomonow

v2.24.0

2.24.0

(Full Changelog)

This release notably introduces a new context command @file:<file-path> to the chat UI, which includes the content of the target file with your prompt when sent. This allows you to ask questions like:

• What does @file:src/components/ActionButton.tsx do?
• Can you refactor @file:src/index.ts to use async/await syntax?
• How do I add an optional dependency to @file:pyproject.toml?

The context command feature also includes an autocomplete menu UI to help navigate your filesystem with fewer keystrokes.

Thank you @michaelchia for developing this feature!

Enhancements made

• Migrate to ChatOllama base class in Ollama provider #1015 (@srdas)
• Add metadata field to agent messages #1013 (@dlqqq)
• Add OpenRouter support #996 (@akaihola)
• Framework for adding context to LLM prompt #993 (@michaelchia)
• Adds unix shell-style wildcard matching to /learn #989 (@andrewfulton9)

Bugs fixed

• Run mypy on CI, fix or ignore typing issues #987 (@krassowski)

Maintenance and upkeep improvements

• Upgrade to actions/upload-artifact@v4 in workflows #992 (@dlqqq)

Contributors to this release

(GitHub contributors page for this release)

@akaihola | @andrewfulton9 | @dlqqq | @ellisonbg | @hockeymomonow | @krassowski | @michaelchia | @srdas

v2.23.0

2.23.0

(Full Changelog)

Enhancements made

• Allow unlimited LLM memory through traitlets configuration #986 (@krassowski)
• Allow to disable automatic inline completions #981 (@krassowski)
• Add ability to delete messages + start new chat session #951 (@michaelchia)

Bugs fixed

• Fix RunnableWithMessageHistory import #980 (@krassowski)
• Fix sort messages #975 (@michaelchia)

Contributors to this release

(GitHub contributors page for this release)

@dlqqq | @krassowski | @michaelchia | @srdas

v2.22.0

2.22.0

(Full Changelog)

Enhancements made

• Add 'Generative AI' submenu #971 (@dlqqq)
• Add Gemini 1.5 to the list of chat options #964 (@trducng)
• Allow configuring a default model for cell magics (and line error magic) #962 (@krassowski)
• Make chat memory size traitlet configurable + /clear to reset memory #943 (@michaelchia)

Maintenance and upkeep improvements

Documentation improvements

• Update documentation to cover installation of all dependencies #961 (@srdas)

Contributors to this release

(GitHub contributors page for this release)

@dlqqq | @krassowski | @michaelchia | @pre-commit-ci | @srdas | @trducng

v2.21.0

2.21.0

(Full Changelog)

Enhancements made

• Add optional configurable message footer #942 (@dlqqq)
• Add support for Azure Open AI Embeddings to Jupyter AI #940 (@gsrikant7)
• Make help message template configurable #938 (@dlqqq)
• Add latest Bedrock models (Titan, Llama 3.1 405b, Mistral Large 2, Jamba Instruct) #923 (@gabrielkoo)
• Add support for custom/provisioned models in Bedrock #922 (@dlqqq)
• Settings section improvement #918 (@andrewfulton9)

Bugs fixed

• Bind reject method to promise, improve typing #949 (@krassowski)
• Fix sending empty input with Enter #946 (@michaelchia)
• Fix saving chat settings #935 (@dlqqq)

Documentation improvements

• Add documentation on how to use Amazon Bedrock #936 (@srdas)
• Update copyright template #925 (@srdas)

Contributors to this release

(GitHub contributors page for this release)

@andrewfulton9 | @dlqqq | @gabrielkoo | @gsrikant7 | @krassowski | @michaelchia | @srdas

v2.20.0

2.20.0

(Full Changelog)

Enhancements made

• Respect selected persona in chat input placeholder #916 (@dlqqq)
• Migrate to langchain-aws for AWS providers #909 (@dlqqq)
• Added new Bedrock Llama 3.1 models and gpt-4o-mini #908 (@srdas)
• Rework selection inclusion; new Send button UX #905 (@dlqqq)

Contributors to this release

(GitHub contributors page for this release)

@dlqqq | @JasonWeill | @srdas

v2.19.1

2.19.1

(Full Changelog)

Enhancements made

• Allow overriding the Ollama base URL #904 (@jtpio)
• Make magic aliases user-customizable #901 (@krassowski)

Bugs fixed

• Trim leading whitespace when processing #900 (@krassowski)
• Fix python<3.10 compatibility #899 (@michaelchia)

Maintenance and upkeep improvements

Documentation improvements

• Add notebooks to the documentation #906 (@andrewfulton9)
• Update docs to reflect Python 3.12 support #898 (@dlqqq)

Contributors to this release

(GitHub contributors page for this release)

@andrewfulton9 | @dlqqq | @jtpio | @krassowski | @michaelchia | @pre-commit-ci

v2.19.0

2.19.0

This is a significant release that implements LLM response streaming in Jupyter AI along with several other enhancements & fixes listed below. Special thanks to @krassowski for his generous contributions this release!

(Full Changelog)

Enhancements made

• Upgrade to langchain~=0.2.0 and langchain_community~=0.2.0 #897 (@dlqqq)
• Rework selection replacement #895 (@dlqqq)
• Ensure all slash commands support -h/--help #878 (@krassowski)
• Add keyboard shortcut command to focus chat input #876 (@krassowski)
• Implement LLM response streaming #859 (@dlqqq)
• Add Ollama #646 (@jtpio)

Bugs fixed

• Fix streaming in HuggingFaceHub provider #894 (@krassowski)
• Fix removal of pending messages on error #888 (@krassowski)
• Ensuring restricted access to the /learn index directory #887 (@krassowski)
• Make preferred-dir the default read/write directory for slash commands #881 (@andrewfulton9)
• Fix prefix removal when streaming inline completions #879 (@krassowski)
• Limit chat input height to 20 lines #877 (@krassowski)
• Do not redefine refreshCompleterState on each render #875 (@krassowski)
• Remove unused toolbars/menus from schema #873 (@krassowski)
• Fix plugin ID format #872 (@krassowski)
• Address error on /learn after change of embedding model #870 (@srdas)
• Fix pending message overlapping text #857 (@michaelchia)
• Fixes error when allowed or blocked model list is passed in config #855 (@3coins)
• Fixed /export for timestamp, agent name #854 (@srdas)

Maintenance and upkeep improvements

• Update to actions/checkout@v4 #893 (@jtpio)
• Upload jupyter-releaser built distributions #892 (@jtpio)
• Updated integration tests workflow #890 (@krassowski)

Contributors to this release

(GitHub contributors page for this release)

@3coins | @andrewfulton9 | @brichet | @dannongruver | @dlqqq | @JasonWeill | @jtpio | @krassowski | @lalanikarim | @michaelchia | @pedrogutobjj | @srdas

v2.18.1

2.18.1

(Full Changelog)

Enhancements made

• Add claude sonnet 3.5 models #847 (@srdas)
• Update clear slash command to use HelpChatHandler to reinstate the help menu #846 (@srdas)

Bugs fixed

• Fix send via keyboard after sending slash command with arguments #850 (@dlqqq)
• Fix Cohere models by using new langchain-cohere partner package #848 (@dlqqq)

Contributors to this release

(GitHub contributors page for this release)

@dlqqq | @srdas