Ollama: lfm3: 1.7B text model with 256K-token context window

What changed

5 points

1. 1

   The lfm3 model is a 1.7B text model with a 256K-token context window, and the lfm3-audio model provides speech recognition, translation, and text-to-speech in 18 languages.

   Official wording · c1 c2

   a 1.7B text model with a 256K-token context window

   speech recognition, translation, and text-to-speech in 18 languages

2. 2

   In this release, lfm3 and d1 2b are supported on Apple hardware, and support for Linux/Windows is coming soon.

   Official wording · c3 c4

   lfm3 and d1 2b are supported on Apple hardware in this release.

   Support for Linux/Windows coming soon.

3. 3

   Liquid AI's d1 models now come in a 2B size, and they are built on LFM3.

   Official wording · c5

   Liquid AI's d1 models now come in a 2B size, built on LFM3.

4. 4

   Speculative decoding now supports DFlash2 draft models when running on MLX, and the ollama launch command now works with OpenCode v2.

   Official wording · c6 c7

   Speculative decoding now supports DFlash2 draft models on MLX

   `ollama launch` now works with OpenCode v2

5. 5

   Prompt processing is now faster on Apple silicon for the Gemma 4, Flint and Laguna models, and a fix was made for tool-only assistant turns from Kimi CLI.

   Official wording · c8 c9

   Faster prompt processing on Apple silicon for Gemma 4, Flint and Laguna models

   Fixed tool-only assistant turns from Kimi CLI

Change details

Fields in this table are taken verbatim from the official text.

Version timeline

Published updates for this tool on this site, newest official date first.

1. 2026-10-27Currentlfm3: 1.7B text model with 256K-token context window
2. 2026-10-21Ollama v0.40.8 adds Flint coding model, supported on MacOS

About this tool

Ollama

Ollama is a tool for running open models locally or in the cloud and building AI applications, with model and documentation resources on its site.

CategoriesAI Models

View Ollama details