LFM3-Omni-2B
LFM3-Omni-2B combines text, image, video and audio understanding with text and speech output.
What the release provides
The 2.17B checkpoint has a shared 32K window and processes a video's images and sound together. It is a non-reasoning model; enable_thinking=True is unsupported.LFM3-Omni-2B official model card
How to use it well
Practical guidance. Review tasks that need more than one input modality, such as matching a spoken event to a video frame. Preserve the timing relationship between sound and images. Keep the model's actual input preparation separate from the user interface's file upload support.
Hosted API contract
The API catalog lists lfm3-omni-2b with a 128,000-token window. Its supported inputs and limits can differ from the downloadable checkpoint.Liquid API model catalog
Specifications & access
| Detail | Value and conditions |
|---|---|
| Access | API, Downloadable |
| Model identifiers | LiquidAI/LFM3-Omni-2Blfm3-omni-2b |
| Context window | 32,768 tokens Downloadable checkpoint limit; the hosted API has a separate service contract. LFM3-Omni-2B official model card |