Quick answer: LFM3-Omni-2B is Liquid AI's first omni-modal model in the LFM3 family (released Oct 27, 2026): a 2.17B-parameter end-to-end model that takes text, images, video (with or without soundtrack), and 16 kHz audio as input and answers in text, 24 kHz speech, or interleaved text and speech. It is designed to replace a separate ASR, vision-language model, LLM, and TTS model on a single device, with a 32K-token context window.
Where LFM3-Omni-2B leads
Where it lags
Best for: on-device voice/vision assistants needing a single small model for speech recognition, image/video Q&A, and spoken responses — e.g., voice chat with visual grounding on phones or laptops.
LFM3-Omni-2B is Liquid AI's first fully omni-modal release in the LFM3 family, pairing the LFM3-1.7B text backbone with an LFM3 vision encoder, a causal audio encoder, and the LiquidCodec audio tokenizer. At 2.17B total parameters, it processes text, images, video (sampled at 2 fps, 4-384 frames, with or without its soundtrack), and 16 kHz audio, and can answer in text, 24 kHz speech, or interleaved text-and-speech depending on the system prompt and generation mode. Liquid frames it as a single model that can replace a separate ASR model, vision-language model, LLM, and TTS model in an on-device pipeline — ASR, TTS, translation, and audio/image/video understanding are all exposed as system-prompt-selected modes on the same checkpoint.
It is explicitly a non-reasoning model (no "thinking" mode), and ships GGUF support for a realtime local speech server via llama.cpp, positioning it for edge voice-assistant deployments rather than complex multi-step agentic work.
| Field | Value |
|---|---|
| Organization | Liquid AI |
| Parameters | 2.17B total |
| Context window | 32,768 tokens (input + output) |
| Architecture | LFM3 backbone + LFM3 vision encoder + LFM3 causal audio encoder + LiquidCodec audio tokenizer |
| License | LFM Open License v1.0 (lfm1.0) |
| Release date | October 27, 2026 |
| Modality | Omni — text, image, video, audio in; text/speech out (16 text languages; 19 speech languages/locales) |
Open weights, self-hostable — no per-token API pricing. Available on Hugging Face in BF16 (SGLang/Transformers) and GGUF (llama.cpp, including a realtime speech server) formats.
| Benchmark | Score | Source | Date |
|---|---|---|---|
| POPE (image+audio) | 83.78 | LFM3-Omni-2B model card | 2026-10 |
| RealWorldQA (spoken) | 64.68 | LFM3-Omni-2B model card | 2026-10 |
| MMStar (spoken) | 51.79 | LFM3-Omni-2B model card | 2026-10 |
| ERQA (spoken) | 39.43 | LFM3-Omni-2B model card | 2026-10 |
| VoiceBench | 64.73 | LFM3-Omni-2B model card | 2026-10 |
| OmniBench (overall) | 46.32 | LFM3-Omni-2B model card | 2026-10 |
Scores above are reported by Liquid AI and shown for context. They are not Benchgen measurements. All scores on a 0-100 scale. See the model card for full evaluation methodology.
Excluded from the table above (bucket-C, documented per skill procedure):
Specs and scores sourced from Liquid AI's official Hugging Face model card (huggingface.co/LiquidAI/LFM3-Omni-2B) and release blog post (liquid.ai/blog/lfm3). Last updated 2026-11-04.