v0.2.13

• Feat #209 : Upgraded to LangChain4j 0.33.0
• Fix #211 : Class initialization must not depend on services
• Feat #213 : Show input/output tokens and cost per request in footer of response

v0.2.12

• Fix #203 : Google WebSeach is broken
• Feat #199 : Show execution time of prompt enhancement
• Fix - Token, Cost and Context Window Settings page mapping correction
• Fix #202 : Update the Gradle IntelliJ build file so it can be installed on other IntelliJ products

v0.2.10

• Fix #184 - Input panel has bigger min/preferred height size
• Feat #186 - Support for local LLaMA.c++ http server
• Feat #191 - Add Google model : gemini-1.5-pro-exp-0801
• Fix #181 - Last selected LLM provider is not persisted anymore, fixed by @mydeveloperplanet
• Feat #181 - Support for multiple projects with different LLM providers & language models
• Fix #190 - Scroll output panel to the bottom when new output is added

v0.2.9

Fix #183 : Allows a remote Ollama instance to be used.

v0.2.8

Support for Exo which allows you to run a local LLM cluster with Llama 3.1 using 8b, 70b and 405b on your Apple Silicon computers.

v0.2.7

• Show window context for downloaded Ollama models
• Also allow Token Calculation + "Add full project" to Ollama models 🔥

v0.2.6

• Renamed Gemini LLM provider to Google
• Increased Gemini Pro 1.5 window context to 2M
• Sorting LLM providers and model names alphabetically in combobox
• LLM cost calculation refactored

v0.2.5

Feat #171: Support OpenAI GPT 4o mini
Fix #170: Fixed LMStudio comms

v0.2.4

Feat #164: Include all attached files in response output reference
Feat #166: Improve code inclusion for chat context

v0.2.3

Feat #148: Create custom commands
Feat #157: Calc tokens for directory
Fix #153: Use the "Copy Project" settings when using "Add Directory to Context Window"
Feat #159: Introduce variable TokenCalculator based on selected LLM Provider
Feat #161: Move predefined command to custom commands