The latest update to the autonomous coding agent Klein, version 3.15, introduces a game-changing feature: the task timeline. This visual storyboard provides a step-by-step replay of how Klein tackles complex coding tasks, offering unprecedented transparency into the agent's inner workings.

The task timeline is displayed in a dedicated panel, where each block represents a key step in the coding process - from the initial prompt to the execution of client-side tools and file edits. By hovering over these blocks, you can instantly review the summary of what happened and when, giving you a clear understanding of Klein's decision-making process.

This feature is particularly crucial as AI coding assistants like Klein often operate with large context windows, which can sometimes lead to performance degradation or missed instructions. The new update introduces intelligent context management, using built-in tracking and customizable client rules to seamlessly transition between tasks while preserving essential information and the full history of previous sessions.

The task timeline ensures high performance by visually highlighting the context limits and how Klein responds to them, providing you with a transparent view of the agent's workflow. This level of visibility is a game-changer, empowering users to better collaborate with the AI and gain deeper insights into its capabilities.

One of the powerful new features introduced in the latest update is the implementation of implicit caching for Gemini 2.5 models. Starting this week, developers no longer need to manually configure cache settings for repeated prompts. Gemini will now automatically apply caching behind the scenes, significantly reducing friction and cost for most users.

While explicit caching will still be available for those who want guaranteed cache hits, the implicit caching feature can save users up to 75% on token costs when reusing parts of their prompts. This is a major step towards making high-quality AI more affordable, especially during long, iterative coding tasks.

Additionally, Google is working on further improvements, such as clear cache hit indicators, refined timeout logic for explicit caching, and a new AI studio usage dashboard. These enhancements will provide users with more insights into their token usage and costs, empowering them to optimize their AI workflows effectively.

With the release of Klein 3.15, the autonomous coding agent has introduced several user experience (UX) improvements to enhance the overall interaction and productivity.

One of the notable improvements is the seamless file editing experience. Previously, when Klein was used to edit files, the interface could become unresponsive, temporarily blocking the user from typing, scrolling, or navigating through the project. This issue has now been resolved, allowing users to seamlessly continue interacting with their IDE while Klein works in the background.

Another valuable feature is the ability to quote messages within the chat interface. This functionality enables more precise conversations and clearer feedback to the AI. Users can now highlight a specific part of their previous message and click the quotation button to reference it within the chat, making it easier to target feedback or request specific actions from Klein.

Additionally, the update has brought a cleaner and more organized interface. The formatting of the chat has been improved, making it easier to manage models and navigate through the various settings. The addition of a robot feature also contributes to the overall aesthetic and readability of the interface.

These UX improvements demonstrate the Klein team's commitment to enhancing the user experience and making the autonomous coding agent more accessible and efficient for developers.

Within the settings tab of the Cline AI assistant, you can access a free model option. This allows you to try out state-of-the-art AI models like CLAW 3.5, Sonnet, and GPT models completely for free.

To access the free model, simply select the "VS Code LM API" option and log in with your GitHub account. This provides you with a way to test and evaluate different AI models without any cost.

However, it's important to note that there is a rate limit on the monthly usage of this free API. It's not an unlimited usage, but rather a way for you to experiment and determine which model works best for your needs before potentially investing in a paid subscription.

By offering this free API access, the Cline team is making it easier for developers to explore and leverage powerful AI capabilities without financial barriers. This can be particularly useful for those looking to create AI-powered applications but want to first assess the available models and their performance.

The latest update to the autonomous coding agent Klein has brought a wealth of new features and improvements that make it one of the best free AI coding assistants available. The introduction of the task timeline, which provides a visual storyboard of the coding process, is a standout feature that offers transparency and helps manage context. The integration of Gemini's implicit caching, which can save up to 75% on token costs, is another significant enhancement. Additionally, the team has made UX improvements, such as seamless file editing and the ability to quote messages within the chat interface. The availability of a free model through the VS Code LM API is a valuable option for those on a budget. With the addition of new model integrations and ongoing bug fixes and improvements, Klein continues to evolve and solidify its position as a powerful AI coding assistant.