# ollama/release
Ollama v0.40.12
importance: highfeature
Adds support for Liquid AI’s LFM3 model family, including text, vision, audio, and multimodal variants, plus 2B d1 decision models. The release also improves Apple silicon prompt processing and image detail, adds MLX DFlash2 speculative decoding support, and includes compatibility and tool-turn fixes.
affected
Users of Liquid AI models and Apple silicon users are affected; LFM3 and d1 2B support is limited to Apple hardware in this release, with Linux and Windows support described as coming soon.
action
Apple hardware users can try the new models with `ollama run lfm3` or `ollama run d1:2b`.
release_signals
+Added support for Liquid AI LFM3 models: lfm3, lfm3-vl, lfm3-audio, and lfm3-omni.
+Added support for Liquid AI d1 and d1-omni decision models in a 2B size.
+Speculative decoding now supports DFlash2 draft models on MLX.
+`ollama launch` now works with OpenCode v2.
code_snippet
ollama run lfm3# community_signal
How useful is this release to your stack?
—
net score
Sign in with GitHub to cast your vote.
Sign in with GitHub to vote
You’ll return to this release, then choose your vote.