Cline v4.1.6 リリースCline Releases v4.1.6

Cline has released version 4.1.6 of its open-source coding agent, a maintenance update that broadens the set of models reachable through the built-in Cline provider and repairs a telemetry bug that could pin run failures on the wrong model. The changes are incremental rather than sweeping, but they matter to developers and teams who depend on accurate per-model reporting when deciding how to route their work.

The most visible change is a new model offering, meta/muse-spark-1.2-contributor, made available on the Cline provider. According to the release notes, it arrives alongside a refreshed model catalog, which indicates the broader list of selectable models has been updated rather than just a single entry being appended. For users, this typically means the model picker inside Cline surfaces new choices and, in some cases, retires or reorders older ones. The release does not specify pricing, availability by region, or benchmark performance for the new model, so those details are best confirmed directly in the client once the update is installed.

The second notable change is a fix to how Cline attributes errors in its telemetry. In earlier builds, an attribute-error report could be logged against a model that was not actually handling a given run, which meant a failure might be counted against the wrong model. Version 4.1.6 corrects this so that failures are reported against the model that was in use for that run. On its own, a mislabeled error may sound minor, but telemetry accuracy compounds over time. If failures are consistently charged to the wrong model, aggregate reliability statistics become misleading, and any decision that leans on them, such as choosing a default model or diagnosing why certain tasks fail, can be skewed. The fix appears aimed at keeping those signals trustworthy.

For readers unfamiliar with the project, Cline is an AI coding assistant that runs as an extension inside editors such as Visual Studio Code. It operates as an agent that can read and edit files, run terminal commands, and work through multi-step tasks with user oversight, rather than functioning purely as an inline autocomplete tool. Because it is model-agnostic, Cline connects to a range of providers, including large commercial APIs and aggregators, and it also offers its own Cline provider as a routing and billing path. Adding a model to that provider makes it selectable without users having to configure a separate account or key, which lowers the barrier to trying different models for coding tasks.

The distinction between a provider and a model is worth keeping in mind. A provider is the endpoint through which requests are sent, while a model is the specific system that generates responses. A single provider can expose many models, and the same underlying model can sometimes be reached through more than one provider. A refreshed catalog on the Cline provider therefore reflects changes to what that particular endpoint offers, and does not necessarily alter models available through other configured providers.

Telemetry, in this context, refers to the diagnostic and usage data the client collects to monitor behavior such as errors and run outcomes. Correctly associating that data with the model responsible is a prerequisite for meaningful analysis, especially in a tool designed to switch between many models. The changelog frames this specifically as fixing attribute-error telemetry, so the scope is the accuracy of error attribution rather than a change to what data is gathered.

The update fits a broader pattern across AI coding tools, where frequent releases juggle model catalog updates with reliability fixes. Cline sits in a competitive field that includes GitHub Copilot, Cursor, and other open-source agents such as Aider, Continue, and Roo Code, a fork of Cline itself. In that landscape, keeping the model lineup current and ensuring instrumentation is trustworthy are both routine but consequential parts of ongoing maintenance.

The project points to a full changelog spanning v4.1.5 to v4.1.6 for the complete list of commits. Users who want the new model or the corrected telemetry behavior will need to update to 4.1.6 through their extension marketplace or the project's distribution channels, and can consult the model picker to confirm which options are now present.