llm-openrouter 0.7.1 release fixes performance issue loading OpenRouter-hosted models

llm-openrouter 0.7.1 is a maintenance release that resolves a performance issue when loading models hosted by OpenRouter. The patch does not add new functionality or expand model compatibility, but it directly addresses a bottleneck in the model-loading process. This suggests the fix is likely narrow in scope, possibly involving API call optimizations or caching improvements rather than a rewrite of the plugin’s core logic.

For engineers integrating OpenRouter’s model catalog into their workflows, this update may reduce latency when initializing or switching between models. The performance gain is specific to OpenRouter-hosted models, so users relying on other providers through the llm plugin will not see changes. The lack of additional features or breaking changes means adoption should be low-risk, though testing in staging environments is still advisable to confirm the fix behaves as expected in your specific setup.

The contribution from waveplate highlights community involvement in maintaining the plugin, which may indicate ongoing interest in OpenRouter’s model ecosystem. However, the single-feed coverage of this release suggests it is a minor update rather than a major shift in the plugin’s capabilities. If your use case depends on OpenRouter’s models, this patch is worth applying, but it does not signal broader changes to the plugin’s roadmap or architecture.