Ollama v0.40.13 Brings LFM3 Model Support to Linux and Windows

Ollama v0.40.13 democratizes access to LiquidAI's LFM3 model family by enabling full support on Linux and Windows systems. Previously, LFM3 models—including the newly-released lfm3-audio with text-to-speech capabilities and lfm3-omni for multimodal tasks—were limited to Apple silicon through MLX. This release removes that platform barrier, allowing Windows and Linux developers to run production-ready local models with audio capabilities without proprietary backends.

This is a significant infrastructure milestone because Ollama is the most widely-adopted local LLM runtime across the industry. By rapidly integrating LFM3 support post-release, the project demonstrates its responsiveness to emerging model architectures and ensures practitioners on all major platforms can access cutting-edge, locally-optimized models without friction. The move also signals that LFM3 is becoming a standard target for local deployment tools.

Read the full article on Ollama release.

Source: Ollama release · Relevance: 8/10