---
title: "Liquid AI models: release history and timeline"
description: "Liquid AI is an efficiency-first foundation model company spun out of MIT CSAIL that builds Liquid Foundation Models (LFMs) for deployment on devices and in the cloud."
type: "release-archive"
canonical_url: "https://theaiengineering.com/releases/liquid-ai"
markdown_url: "https://theaiengineering.com/releases/liquid-ai.md"
author: "The AI Engineering Editors"
publisher: "The AI Engineering"
updated: "2026-10-27"
citation: "The AI Engineering Editors. \"Liquid AI models: release history and timeline.\" The AI Engineering, updated Oct 27, 2026. https://theaiengineering.com/releases/liquid-ai"
license: "CC BY 4.0"
license_url: "https://creativecommons.org/licenses/by/4.0/"
data_csv: "https://theaiengineering.com/releases.csv"
data_json: "https://theaiengineering.com/releases.json"
---

# Liquid AI models: release history and timeline

Liquid AI is an efficiency-first foundation model company spun out of MIT CSAIL that builds Liquid Foundation Models (LFMs) for deployment on devices and in the cloud.

- Headquarters: Cambridge, United States
- Founded: 2023
- Website: https://www.liquid.ai

34 releases from 1 lab, 30 with open weights. Dates are the labs' official announcement dates.

## October 2026

### d1-2B and d1-Omni-2B

Two open-weight LFM3-based decision models: d1-2B, built on LFM3-VL-2B for text, image, and video inputs, and d1-Omni-2B, built on LFM3-Omni-2B with audio input, adding typed answers such as boxes, values, and estimates.

- Date: 2026-10-27
- Lab: Liquid AI
- Type: Small model
- Weights: Open weights
- Modalities: text, image, video, audio → score
- Parameters: d1-2B: 1.92B; d1-Omni-2B: 2.17B
- License: LFM Open License v1.0
- Available at launch: Open weights
- Permalink: https://theaiengineering.com/releases/2026/10#d1-2b-omni-2b

Highlights:

- d1-2B replaces d1-3B with about a third fewer parameters
- Typed answers: point (boxes), read (text, numbers, dates), estimate, and multi-select
- Audio input is d1-Omni-2B only; 128K-token context per model cards
- Weights on Hugging Face under the LFM Open License

Source: [Introducing LFM3: Every Modality for Specialized Intelligence](https://www.liquid.ai/blog/lfm3)

### LFM3-1.7B

The text model of Liquid AI's LFM3 generation, built on a new tri-hybrid architecture of gated short convolutions, sliding-window attention, and global attention, pre-trained on 22T tokens and post-trained for local agent runtimes.

- Date: 2026-10-27
- Lab: Liquid AI
- Type: Small model
- Weights: Open weights
- Modalities: text → text
- Context window: 262,144 tokens
- Parameters: 1.73B
- License: LFM Open License v1.0
- Available at launch: API, Consumer app, Open weights
- Permalink: https://theaiengineering.com/releases/2026/10#lfm3-1-7b

Highlights:

- 256K-token context window (262,144 tokens per model card)
- 24 layers: 13 gated short convolutions, 5 sliding-window attention, 6 global attention
- Post-trained for agent runtimes such as Hermes Agent, Pi, and OpenClaw
- Open weights on Hugging Face; Liquid API model ID lfm3-1.7b

Source: [Introducing LFM3: Every Modality for Specialized Intelligence](https://www.liquid.ai/blog/lfm3)

### LFM3-Audio-2B

A 2B audio model in the LFM3 family with an in-house audio encoder and codec that takes text or speech and replies with text, speech, or both, adding audio understanding, reasoning, and tool calling.

- Date: 2026-10-27
- Lab: Liquid AI
- Type: Speech
- Weights: Open weights
- Modalities: text, audio → text, audio
- Parameters: 2B
- License: LFM Open License v1.0
- Available at launch: API, Consumer app, Open weights
- Permalink: https://theaiengineering.com/releases/2026/10#lfm3-audio-2b

Highlights:

- Voice chat, transcription, speech synthesis, and speech translation in one model
- Speech support expanded to 18 languages
- Outputs 24 kHz speech; audio encoder operates at 16 kHz
- Liquid API model ID lfm3-audio-2b; open weights on Hugging Face

Source: [Introducing LFM3: Every Modality for Specialized Intelligence](https://www.liquid.ai/blog/lfm3)

### LFM3-Omni-2B

Liquid AI's first omni model, combining vision, audio, and language in one LFM3-based model that accepts text, images, video with sound, and speech, and answers in text, speech, or both.

- Date: 2026-10-27
- Lab: Liquid AI
- Type: Multimodal
- Weights: Open weights
- Modalities: text, image, video, audio → text, audio
- Parameters: 2.17B
- License: LFM Open License v1.0
- Available at launch: API, Consumer app, Open weights
- Permalink: https://theaiengineering.com/releases/2026/10#lfm3-omni-2b

Highlights:

- Supports spoken visual question answering and audio-visual video understanding
- Tool calling from combined visual context and spoken instructions
- Liquid API model ID lfm3-omni-2b; open weights on Hugging Face

Source: [Introducing LFM3: Every Modality for Specialized Intelligence](https://www.liquid.ai/blog/lfm3)

### LFM3-VL-2B

The LFM3 vision-language model, branched from LFM3-1.7B-Base and the first LFM vision model to accept video input, adding step-by-step thinking and spatial reasoning for localizing objects.

- Date: 2026-10-27
- Lab: Liquid AI
- Type: Multimodal
- Weights: Open weights
- Modalities: text, image, video → text
- Context window: 128,000 tokens
- Parameters: 1.92B
- License: LFM Open License v1.0
- Available at launch: API, Consumer app, Open weights
- Permalink: https://theaiengineering.com/releases/2026/10#lfm3-vl-2b

Highlights:

- 1.92B parameters: 1.73B language model, 164M vision encoder, 25M projector
- Video input sampled at 2 fps per the model card
- Liquid API model ID lfm3-vl-2b with a 128,000-token context window
- Open weights on Hugging Face

Source: [Introducing LFM3: Every Modality for Specialized Intelligence](https://www.liquid.ai/blog/lfm3)

### d1-ultra

A hosted multimodal decision model that Liquid AI built on a larger open-weight base than d1 with Liquid-specific customizations, trained to produce better-calibrated probabilities for harder text and image decisions.

- Date: 2026-10-15
- Lab: Liquid AI
- Type: Multimodal
- Weights: Proprietary
- Modalities: text, image → score
- Context window: 32,768 tokens
- Available at launch: API
- Permalink: https://theaiengineering.com/releases/2026/10#d1-ultra

Highlights:

- Liquid API model ID d1-ultra on the decisions endpoint
- Returns Noul, Choice, and Score answers without generating tokens
- 32,768-token context window per Liquid API docs
- Later listed on OpenRouter and Vercel AI Gateway per Liquid docs

Source: [d1-ultra: Introducing the World’s Best Decision Model](https://www.liquid.ai/blog/d1-ultra)

### LFM2.5-24B-A2B

The largest and final LFM2.5 model, a reasoning-only mixture-of-experts model with 24B total and about 2B active parameters, extending LFM2-24B-A2B with a 131K context window, larger vocabulary, and agentic reinforcement learning.

- Date: 2026-10-13
- Lab: Liquid AI
- Type: Reasoning
- Weights: Open weights
- Modalities: text → text
- Context window: 131,072 tokens
- Parameters: MoE: 24.0B total, 2.45B active
- License: LFM Open License v1.0
- Available at launch: API, Consumer app, Open weights
- Permalink: https://theaiengineering.com/releases/2026/10#lfm2-5-24b-a2b

Highlights:

- Context extended from 32,768 to 131,072 tokens; vocabulary doubled from 65K to 128K
- Pre-training scaled from 17T to 33T tokens
- Agentic RL inside agent runtimes such as OpenClaw, Hermes Agent, Pi, and OpenCode
- Base and post-trained weights on Hugging Face; also on Liquid Playground and Liquid API

Source: [LFM2.5-24B-A2B: Built for Fast On-Device Agents](https://www.liquid.ai/blog/lfm2-5-24b-a2b)

### d1-3B and d1-omni-600M

The first open-weight d1 decision models: d1-3B, trained from LFM2.5-VL-3B for text and image inputs, and the experimental d1-omni-600M, built on LFM2.5-Encoder-350M with added vision and audio encoders.

- Date: 2026-10-07
- Lab: Liquid AI
- Type: Small model
- Weights: Open weights
- Modalities: text, image, audio → score
- Parameters: d1-3B: 3.12B; d1-omni-600M: 587M
- License: LFM Open License v1.0
- Available at launch: Open weights
- Permalink: https://theaiengineering.com/releases/2026/10#d1-3b-omni-600m

Highlights:

- Answers questions in one forward pass without generating tokens
- d1-omni-600M accepts text with images or text with audio
- Context: 32,768 tokens (d1-3B) and 16,384 tokens (d1-omni-600M) per model cards
- Available on Hugging Face with day-one llama.cpp support

Source: [Open d1: Edge decision models for text, vision, and audio](https://www.liquid.ai/blog/d1-open)

### d1

Liquid AI's first decision model, which reads text and images with one or more questions in a single forward pass and returns probabilities for each answer instead of generating tokens; this release adds image input.

- Date: 2026-10-05
- Lab: Liquid AI
- Type: Multimodal
- Weights: Proprietary
- Modalities: text, image → score
- Available at launch: API
- Permalink: https://theaiengineering.com/releases/2026/10#d1-decision-model

Highlights:

- Liquid API model ID d1, served through the decisions endpoint
- Question types: Noul (yes/no), Choice, and Score
- Also on OpenRouter and Vercel AI Gateway, text only at launch
- Follows an experimental text-only d1 release the prior week

Source: [Introducing d1: The most capable decision model, now with vision](https://www.liquid.ai/blog/d1-decision-model)

## September 2026

### LFM2-1.2B-Longevity and LFM2-2.6B-Longevity

Domain-adapted LFM2 models fine-tuned on aging-biology data spanning clinical records, DNA methylation, transcriptomics, proteomics, and genetics, released with the LongevityBench benchmark and an accompanying Cell paper.

- Date: 2026-09-17
- Lab: Liquid AI
- Type: Small model
- Weights: Open weights
- Modalities: text → text
- Context window: 32,768 tokens
- Parameters: 1.2B; 2.6B
- License: LFM Open License v1.0
- Available at launch: Open weights
- Permalink: https://theaiengineering.com/releases/2026/09#lfm2-longevity

Highlights:

- Fine-tuned from LFM2-1.2B and LFM2-2.6B for three epochs at a 32k-token context
- LongevityBench covers 17 tasks and 25,457 prompts across five biodata domains
- Weights available on Hugging Face

Source: [Setting the Frontier of Aging Biology with Liquid Foundation Models](https://www.liquid.ai/blog/longevitybench)

## August 2026

### LFM2.5-VL-3B

An updated 3B vision-language model building on LFM2-VL-3B with improved screen and UI understanding, grounding, function calling for text and image inputs, and multi-image input, answering directly without a reasoning phase.

- Date: 2026-08-12
- Lab: Liquid AI
- Type: Multimodal
- Weights: Open weights
- Modalities: text, image → text
- Context window: 32,768 tokens
- Parameters: 3.1B
- License: LFM Open License v1.0
- Available at launch: Consumer app, Open weights
- Permalink: https://theaiengineering.com/releases/2026/08#lfm2-5-vl-3b

Highlights:

- Built on the LFM2.5-2.6B-Base backbone with a 128,000-token vocabulary
- Vision pretraining scaled by 4x tokens
- 32,768-token context length per model card
- Available on Hugging Face and Liquid Playground

Source: [LFM2.5-VL-3B: A Better and Faster Vision-Language Model for the Edge](https://www.liquid.ai/blog/lfm2-5-vl-3b)

### LFM2.5-2.6B

An agentic 2.6B model that runs on-device, pre-trained on about 34T tokens with a 128K context extension, a vocabulary doubled to 128K, and a four-stage post-training pipeline ending in agentic reinforcement learning.

- Date: 2026-08-04
- Lab: Liquid AI
- Type: Small model
- Weights: Open weights
- Modalities: text → text
- Context window: 131,072 tokens
- Parameters: 2.6B
- License: LFM Open License v1.0
- Available at launch: Consumer app, Open weights
- Permalink: https://theaiengineering.com/releases/2026/08#lfm2-5-2-6b

Highlights:

- 131,072-token context length per model card
- Post-training: SFT, teacher specialization, multi-domain on-policy distillation, agentic RL
- Base (LFM2.5-2.6B-Base) and post-trained checkpoints on Hugging Face

Source: [LFM2.5-2.6B: Deploy Agents Everywhere](https://www.liquid.ai/blog/lfm2-5-2-6b)

## July 2026

### LFM2.5-Encoder-230M and LFM2.5-Encoder-350M

Two bidirectional masked-language encoders on the LFM2 hybrid architecture, initialized from LFM2.5-230M and LFM2.5-350M and meant to be fine-tuned for classification, natural language understanding, token-level tasks, and retrieval.

- Date: 2026-07-28
- Lab: Liquid AI
- Type: Embedding
- Weights: Open weights
- Modalities: text → embedding
- Context window: 8,192 tokens
- Parameters: 230M; 350M
- License: LFM Open License v1.0
- Available at launch: Open weights
- Permalink: https://theaiengineering.com/releases/2026/07#lfm2-5-encoders

Highlights:

- Native 8,192-token context length
- Trained with a masked language modeling objective that masks 30% of tokens
- Demo fine-tunes include PII detection, spell checking, and a policy linter
- Available on Hugging Face

Source: [LFM2.5-Encoders: Fast at Long Context, Even on CPU](https://www.liquid.ai/blog/lfm2-5-encoders)

## June 2026

### LFM2.5-230M

Liquid AI's smallest model at release, a 230M-parameter LFM2-architecture model pre-trained on 19T tokens with a 32K context extension and a light post-training recipe meant as a base for fine-tuning in agentic workflows.

- Date: 2026-06-25
- Lab: Liquid AI
- Type: Small model
- Weights: Open weights
- Modalities: text → text
- Context window: 32,768 tokens
- Parameters: 230M
- License: LFM Open License v1.0
- Available at launch: Open weights
- Permalink: https://theaiengineering.com/releases/2026/06#lfm2-5-230m

Highlights:

- Released as LFM2.5-230M-Base and post-trained LFM2.5-230M
- Post-training: distillation from LFM2.5-350M, DPO, and multi-domain RL
- Demonstrated as a skill-selection layer on a Unitree G1 humanoid robot
- Available on Hugging Face

Source: [LFM2.5-230M: Built to Run Anywhere](https://www.liquid.ai/blog/lfm2-5-230m)

### LFM2.5-Embedding-350M and LFM2.5-ColBERT-350M

Two multilingual retrieval models, the first bidirectional members of the LFM family: a single-vector embedding model and a late-interaction ColBERT model, both 350M parameters and built from LFM2.5-350M-Base.

- Date: 2026-06-18
- Lab: Liquid AI
- Type: Embedding
- Weights: Open weights
- Modalities: text → embedding
- Parameters: 350M each
- License: LFM Open License v1.0
- Available at launch: Open weights
- Permalink: https://theaiengineering.com/releases/2026/06#lfm2-5-retrievers

Highlights:

- Supports 11 languages for multilingual and cross-lingual search
- Causal attention and short convolutions converted to bidirectional versions
- Positioned for short-context search; 512-token document length per model card
- Available on Hugging Face

Source: [LFM2.5 Retrievers: Bi-directional LFMs for Fast Multilingual Search](https://www.liquid.ai/blog/lfm2-5-retrievers)

## May 2026

### LFM2.5-8B-A1B

An updated on-device mixture-of-experts model built for tool calling, expanding LFM2-8B-A1B with a 128K context window, pretraining scaled from 12T to 38T tokens, a doubled vocabulary, and reasoning-only output.

- Date: 2026-05-28
- Lab: Liquid AI
- Type: Small model
- Weights: Open weights
- Modalities: text → text
- Context window: 128,000 tokens
- Parameters: MoE: 8.3B total, 1.5B active
- License: LFM Open License v1.0
- Available at launch: Consumer app, Open weights
- Permalink: https://theaiengineering.com/releases/2026/05#lfm2-5-8b-a1b

Highlights:

- Context window expanded from 32,768 to 128,000 tokens
- Vocabulary grows from 65,536 to 128,000 tokens for non-Latin scripts
- Produces an explicit chain of thought before every final answer
- Base and post-trained models on Hugging Face; testable on Liquid Playground

Source: [LFM2.5-8B-A1B: An Even Better On-Device Mixture of Experts](https://www.liquid.ai/blog/lfm2-5-8b-a1b)

## April 2026

### LFM2.5-VL-450M

An updated 450M vision-language model that adds bounding-box object detection, function calling support, and better instruction following over LFM2-VL-450M, with pretraining scaled from 10T to 28T tokens.

- Date: 2026-04-08
- Lab: Liquid AI
- Type: Multimodal
- Weights: Open weights
- Modalities: text, image → text
- Context window: 32,768 tokens
- Parameters: 450M
- License: LFM Open License v1.0
- Available at launch: Consumer app, Open weights
- Permalink: https://theaiengineering.com/releases/2026/04#lfm2-5-vl-450m

Highlights:

- Improved multilingual image understanding across eight languages
- Built on the LFM2.5-350M backbone per model card
- Available on Hugging Face, LEAP, and Liquid Playground

Source: [LFM2.5-VL-450M: Structured Visual Intelligence, Edge to Cloud](https://www.liquid.ai/blog/lfm2-5-vl-450m)

## March 2026

### LFM2.5-350M

An updated 350M-parameter model with pretraining extended from 10T to 28T tokens and large-scale reinforcement learning, aimed at tool use, data extraction, and structured outputs on CPUs, NPUs, and GPUs.

- Date: 2026-03-31
- Lab: Liquid AI
- Type: Small model
- Weights: Open weights
- Modalities: text → text
- Context window: 32,768 tokens
- Parameters: 350M
- License: LFM Open License v1.0
- Available at launch: Consumer app, Open weights
- Permalink: https://theaiengineering.com/releases/2026/03#lfm2-5-350m

Highlights:

- Released as LFM2.5-350M-Base and post-trained LFM2.5-350M
- Available on Hugging Face, LEAP, and Liquid Playground
- New ecosystem partners at launch include Zetic, RunAnywhere, and Mirai

Source: [LFM2.5-350M: No Size Left Behind](https://www.liquid.ai/blog/lfm2-5-350m-no-size-left-behind)

## February 2026

### LFM2-24B-A2B

An early checkpoint of the largest LFM2 model, a sparse mixture-of-experts instruct model with 24B total and 2.3B active parameters, designed to fit in 32GB of RAM for laptops, desktops, and cloud deployment.

- Date: 2026-02-24
- Lab: Liquid AI
- Type: LLM
- Weights: Open weights
- Modalities: text → text
- Context window: 32,768 tokens
- Parameters: MoE: 24B total, 2.3B active
- License: LFM Open License v1.0
- Available at launch: Consumer app, Open weights
- Permalink: https://theaiengineering.com/releases/2026/02#lfm2-24b-a2b

Highlights:

- 40 layers with 64 experts per MoE block and top-4 routing
- Instruct model without reasoning traces; trained on 17T tokens at release
- Day-zero support in llama.cpp, vLLM, and SGLang
- Weights on Hugging Face; testable on Liquid Playground

Source: [LFM2-24B-A2B: Scaling Up the LFM2 Architecture](https://www.liquid.ai/blog/lfm2-24b-a2b)

## January 2026

### LFM2.5-1.2B-Thinking

A 1.2B-parameter reasoning model in the LFM2.5 family that generates thinking traces before answering and runs on-device within 900 MB of memory on a phone.

- Date: 2026-01-20
- Lab: Liquid AI
- Type: Reasoning
- Weights: Open weights
- Modalities: text → text
- Context window: 32,768 tokens
- Parameters: 1.17B
- License: LFM Open License v1.0
- Available at launch: Consumer app, Open weights
- Permalink: https://theaiengineering.com/releases/2026/01#lfm2-5-1-2b-thinking

Highlights:

- Available on Hugging Face, LEAP, and Liquid Playground
- New launch partners: Qualcomm Technologies, Ollama, FastFlowLM, and Cactus Compute
- 32,768-token context window per model card

Source: [LFM2.5-1.2B-Thinking: On-Device Reasoning Under 1GB](https://www.liquid.ai/blog/lfm2-5-1-2b-thinking-on-device-reasoning-under-1gb)

### LFM2-2.6B-Transcript

A Liquid Nano fine-tuned from LFM2-2.6B for summarizing 30 to 60 minute meeting transcripts on-device, developed with AMD and showcased at CES running locally across Ryzen AI CPUs, GPUs, and NPUs.

- Date: 2026-01-07
- Lab: Liquid AI
- Type: Small model
- Weights: Open weights
- Modalities: text → text
- Parameters: 2.6B
- License: LFM Open License v1.0
- Available at launch: Open weights
- Permalink: https://theaiengineering.com/releases/2026/01#lfm2-2-6b-transcript

Highlights:

- Outputs structured summaries with key points, decisions, and attributed action items
- Under 3 GB RAM usage reported for the AMD deployment
- Available for download on Hugging Face and LEAP

Source: [The Future of Meeting Summarization: Local, Fast, Private, and Fully Secure](https://www.liquid.ai/blog/the-future-of-meeting-summarization-local-fast-private-and-fully-secure)

### LFM2.5-1.2B (Base, Instruct, JP)

The first LFM2.5 release, a 1.2B text family on the LFM2 architecture with pretraining extended from 10T to 28T tokens and expanded reinforcement learning, shipped as Base, Instruct, and Japanese-optimized checkpoints.

- Date: 2026-01-05
- Lab: Liquid AI
- Type: Small model
- Weights: Open weights
- Modalities: text → text
- Context window: 32,768 tokens
- Parameters: 1.17B
- License: LFM Open License v1.0
- Available at launch: Consumer app, Open weights
- Permalink: https://theaiengineering.com/releases/2026/01#lfm2-5-1-2b

Highlights:

- Checkpoints: LFM2.5-1.2B-Base, LFM2.5-1.2B-Instruct, and LFM2.5-1.2B-JP
- Announced alongside LFM2.5-VL-1.6B and LFM2.5-Audio-1.5B
- Open weights on Hugging Face and LEAP; Instruct also on Liquid Playground
- AMD and Nexa AI provide NPU-optimized versions

Source: [Introducing LFM2.5: The Next Generation of On-Device AI](https://www.liquid.ai/blog/introducing-lfm2-5-the-next-generation-of-on-device-ai)

### LFM2.5-Audio-1.5B

A native audio-language model in the LFM2.5 family that accepts and produces both speech and text, introducing a custom LFM-based audio detokenizer that Liquid AI says is 8x faster than LFM2's Mimi detokenizer.

- Date: 2026-01-05
- Lab: Liquid AI
- Type: Speech
- Weights: Open weights
- Modalities: text, audio → text, audio
- Context window: 32,768 tokens
- Parameters: 1.5B
- License: LFM Open License v1.0
- Available at launch: Consumer app, Open weights
- Permalink: https://theaiengineering.com/releases/2026/01#lfm2-5-audio-1-5b

Highlights:

- llama.cpp-compatible GGUF checkpoints for CPU inference
- Improved ASR and TTS quality over LFM2-Audio-1.5B per Liquid AI
- Available on Hugging Face, LEAP, and Liquid Playground

Source: [Introducing LFM2.5: The Next Generation of On-Device AI](https://www.liquid.ai/blog/introducing-lfm2-5-the-next-generation-of-on-device-ai)

### LFM2.5-VL-1.6B

A refreshed 1.6B vision-language model on the updated LFM2.5 base backbone, released with the LFM2.5 family and tuned for multi-image comprehension, multilingual visual understanding, and instruction following.

- Date: 2026-01-05
- Lab: Liquid AI
- Type: Multimodal
- Weights: Open weights
- Modalities: text, image → text
- Context window: 32,768 tokens
- Parameters: 1.6B
- License: LFM Open License v1.0
- Available at launch: Consumer app, Open weights
- Permalink: https://theaiengineering.com/releases/2026/01#lfm2-5-vl-1-6b

Highlights:

- Handles prompts in Arabic, Chinese, French, German, Japanese, Korean, and Spanish
- Succeeds LFM2-VL-1.6B
- Available on Hugging Face, LEAP, and Liquid Playground

Source: [Introducing LFM2.5: The Next Generation of On-Device AI](https://www.liquid.ai/blog/introducing-lfm2-5-the-next-generation-of-on-device-ai)

## October 2025

### LFM2-ColBERT-350M

A multilingual late-interaction retriever built on the LFM2 backbone that encodes queries and documents at the token level, supporting cross-lingual search where documents in one language are retrieved with queries in another.

- Date: 2025-10-28
- Lab: Liquid AI
- Type: Embedding
- Weights: Open weights
- Modalities: text → embedding
- Parameters: 353M
- License: LFM Open License v1.0
- Available at launch: Open weights
- Permalink: https://theaiengineering.com/releases/2025/10#lfm2-colbert-350m

Highlights:

- Document length 512 tokens and query length 32 tokens per model card
- Positioned as a drop-in replacement in existing RAG pipelines
- Available on Hugging Face with an interactive demo
- Succeeded by LFM2.5-ColBERT-350M in 2026

Source: [LFM2-ColBERT-350M: One Model to Embed Them All](https://www.liquid.ai/blog/lfm2-colbert-350m-one-model-to-embed-them-all)

### LFM2-VL-3B

A 3B-parameter vision-language model built on the LFM2-2.6B backbone with a SigLIP2 400M NaFlex encoder, adding more capacity to the LFM2-VL family and extending visual understanding to ten languages.

- Date: 2025-10-22
- Lab: Liquid AI
- Type: Multimodal
- Weights: Open weights
- Modalities: text, image → text
- Context window: 32,768 tokens
- Parameters: 3B
- License: LFM Open License v1.0
- Available at launch: Open weights
- Permalink: https://theaiengineering.com/releases/2025/10#lfm2-vl-3b

Highlights:

- Processes images at native resolution with variable aspect ratios
- Adjustable vision tokens per image to balance speed and quality
- Available on Hugging Face under the LFM Open License and on LEAP

Source: [LFM2-VL-3B: A New Efficient Vision-Language for the Edge](https://www.liquid.ai/blog/lfm2-vl-3b-a-new-efficient-vision-language-for-the-edge)

### LFM2-8B-A1B

Liquid AI's first on-device mixture-of-experts model, an early checkpoint that adds sparse MoE feed-forward layers to the LFM2 backbone, with 8.3B total and 1.5B active parameters per token.

- Date: 2025-10-07
- Lab: Liquid AI
- Type: Small model
- Weights: Open weights
- Modalities: text → text
- Context window: 32,768 tokens
- Parameters: MoE: 8.3B total, 1.5B active
- License: LFM Open License v1.0
- Available at launch: Consumer app, Open weights
- Permalink: https://theaiengineering.com/releases/2025/10#lfm2-8b-a1b

Highlights:

- 32 experts per MoE block with top-4 routing; 18 convolution and 6 GQA blocks
- Pre-trained on about 12T tokens with more code than LFM2-2.6B
- Weights and GGUF quants on Hugging Face; testable on Liquid Playground
- Later replaced by LFM2.5-8B-A1B (May 2026)

Source: [LFM2-8B-A1B: An Efficient On-device Mixture-of-Experts](https://www.liquid.ai/blog/lfm2-8b-a1b-an-efficient-on-device-mixture-of-experts)

### LFM2-Audio-1.5B

An end-to-end audio and text foundation model that extends LFM2-1.2B to accept and generate both speech and text, covering conversational chat, speech recognition, text-to-speech, and audio classification in one 1.5B-parameter model.

- Date: 2025-10-01
- Lab: Liquid AI
- Type: Speech
- Weights: Open weights
- Modalities: text, audio → text, audio
- Context window: 32,768 tokens
- Parameters: 1.5B
- License: LFM Open License v1.0
- Available at launch: Consumer app, Open weights
- Permalink: https://theaiengineering.com/releases/2025/10#lfm2-audio-1-5b

Highlights:

- Continuous embeddings for audio input; discrete audio tokens for output
- Tokenizer-free input: raw waveform chunked into segments of about 80 ms
- Supported language per model card: English
- Available on Hugging Face and Liquid Playground

Source: [LFM2-Audio: An End-to-End Audio Foundation Model](https://www.liquid.ai/blog/lfm2-audio-an-end-to-end-audio-foundation-model)

## September 2025

### Liquid Nanos (LFM2-Extract, LFM2-350M-ENJP-MT, LFM2-1.2B-RAG, LFM2-1.2B-Tool, LFM2-350M-Math)

A first set of experimental task-specific LFM2 models for on-device agentic building blocks: data extraction, English-Japanese translation, retrieval-augmented question answering, tool calling, and math reasoning, released as open-weight 350M and 1.2B checkpoints.

- Date: 2025-09-25
- Lab: Liquid AI
- Type: Small model
- Weights: Open weights
- Modalities: text → text
- Parameters: 350M and 1.2B checkpoints
- License: LFM Open License v1.0
- Available at launch: Open weights
- Permalink: https://theaiengineering.com/releases/2025/09#lfm2-liquid-nanos

Highlights:

- Models: LFM2-350M-Extract, LFM2-1.2B-Extract, LFM2-350M-ENJP-MT, LFM2-1.2B-RAG, LFM2-1.2B-Tool, LFM2-350M-Math
- Extract models turn unstructured documents into JSON, XML, or YAML
- Available on Hugging Face and Liquid's LEAP platform for iOS, Android, and laptops

Source: [Introducing Liquid Nanos: frontier-grade performance on everyday devices](https://www.liquid.ai/blog/introducing-liquid-nanos-frontier-grade-performance-on-everyday-devices)

### LFM2-2.6B

The largest dense model in the LFM2 series at its release, a 2.6B-parameter hybrid convolution and grouped query attention model trained on 10 trillion tokens and tuned for English and Japanese.

- Date: 2025-09-23
- Lab: Liquid AI
- Type: Small model
- Weights: Open weights
- Modalities: text → text
- Context window: 32,768 tokens
- Parameters: 2.6B
- License: LFM Open License v1.0
- Available at launch: Open weights
- Permalink: https://theaiengineering.com/releases/2025/09#lfm2-2-6b

Highlights:

- Also supports French, Spanish, German, Italian, Portuguese, Arabic, Chinese, and Korean
- 32,768-token context window per model card
- Released on Hugging Face under the LFM Open License

Source: [Introducing LFM2-2.6B: Redefining Efficiency in Language Models](https://www.liquid.ai/blog/introducing-lfm2-2-6b-redefining-efficiency-in-language-models)

## August 2025

### LFM2-VL (450M, 1.6B)

Liquid AI's first vision-language models, extending LFM2 with SigLIP2 NaFlex image encoders to process text and images at native resolution, released as open-weight 450M and 1.6B variants for on-device use.

- Date: 2025-08-12
- Lab: Liquid AI
- Type: Multimodal
- Weights: Open weights
- Modalities: text, image → text
- Context window: 32,768 tokens
- Parameters: 450M; 1.6B
- License: LFM Open License v1.0
- Available at launch: Open weights
- Permalink: https://theaiengineering.com/releases/2025/08#lfm2-vl

Highlights:

- Native image resolution up to 512x512; larger images split into 512x512 patches
- Built on LFM2-350M (450M model) and LFM2-1.2B (1.6B model) backbones
- User-tunable image token count trades speed for quality at inference
- Weights on Hugging Face; supported by LEAP for edge deployment

Source: [LFM2-VL: Efficient Vision-Language Models](https://www.liquid.ai/blog/lfm2-vl-efficient-vision-language-models)

## July 2025

### LFM2 (350M, 700M, 1.2B)

Second-generation Liquid Foundation Models using a new hybrid architecture of gated short convolutions and grouped query attention, released as open-weight 350M, 700M, and 1.2B dense checkpoints designed for fast on-device inference.

- Date: 2025-07-10
- Lab: Liquid AI
- Type: Small model
- Weights: Open weights
- Modalities: text → text
- Context window: 32,768 tokens
- Parameters: 350M: 354M; 700M: 742M; 1.2B: 1.17B
- License: LFM Open License v1.0
- Available at launch: API, Consumer app, Open weights
- Permalink: https://theaiengineering.com/releases/2025/07#lfm2

Highlights:

- Architecture: 10 double-gated short convolution blocks and 6 grouped query attention blocks
- Trained on 10T tokens with LFM1-7B as a knowledge-distillation teacher
- Weights on Hugging Face; also on Liquid Playground and OpenRouter
- LFM Open License v1.0 allows free commercial use below a company revenue threshold

Source: [Introducing LFM2: The Fastest On-Device Foundation Models on the Market](https://www.liquid.ai/blog/liquid-foundation-models-v2-our-second-series-of-generative-ai-models)

## January 2025

### LFM-7B

A 7.7B-parameter Liquid Foundation Model tuned for chat, instruction following, and multilingual use, with Liquid AI highlighting English, Arabic, and Japanese, aimed at private enterprise, on-premises, and on-device deployment.

- Date: 2025-01-20
- Lab: Liquid AI
- Type: LLM
- Weights: Proprietary
- Modalities: text → text
- Parameters: 7.7B
- Available at launch: API, Consumer app, Cloud platforms, Enterprise
- Permalink: https://theaiengineering.com/releases/2025/01#lfm-7b

Highlights:

- Supports English, Spanish, French, German, Chinese, Arabic, Japanese, and Korean
- Effective context length of 32k tokens validated with RULER
- Offered on Liquid Playground, OpenRouter, and AWS Marketplace
- Weights not released; on-premises customization stack sold to enterprises

Source: [Introducing LFM-7B: Setting New Standards for Efficient Language Models](https://www.liquid.ai/blog/introducing-lfm-7b-setting-new-standards-for-efficient-language-models)

## September 2024

### LFM-1B, LFM-3B, and LFM-40B

Liquid AI's first series of Liquid Foundation Models, non-transformer language models built from dynamical systems and signal processing principles, released as dense 1.3B and 3.1B models plus a 40.3B mixture-of-experts model.

- Date: 2024-09-30
- Lab: Liquid AI
- Type: LLM
- Weights: Proprietary
- Modalities: text → text
- Parameters: 1B: 1.3B; 3B: 3.1B; 40B: 40.3B MoE (12B active)
- Available at launch: Consumer app, Enterprise
- Permalink: https://theaiengineering.com/releases/2024/09#lfm-1b-3b-40b

Highlights:

- Preview release optimized for a 32k-token context length
- LFM-40B activates 12B parameters per token
- Available on Liquid Playground; weights not open-sourced
- Liquid AI now marks this generation as deprecated

Source: [Liquid Foundation Models: Our First Series of Generative AI Models](https://www.liquid.ai/blog/liquid-foundation-models-our-first-series-of-generative-ai-models)

## Cite and reuse this data

These release records are free to reuse under CC BY 4.0, including commercially. Credit The AI Engineering and link to this page's URL. Dates are the labs' official announcement dates, and every record links to its source. List fields in the CSV are separated by " | ".

The AI Engineering Editors. "Liquid AI models: release history and timeline." The AI Engineering, updated Oct 27, 2026. https://theaiengineering.com/releases/liquid-ai

HTML link:

```html
<a href="https://theaiengineering.com/releases/liquid-ai">Liquid AI models: release history and timeline</a> (The AI Engineering)
```

- Title: Liquid AI models: release history and timeline
- Author: The AI Engineering Editors
- Publisher: The AI Engineering
- Updated: 2026-10-27
- URL: https://theaiengineering.com/releases/liquid-ai
- Markdown: https://theaiengineering.com/releases/liquid-ai.md
- License: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/)
- Full dataset as CSV: https://theaiengineering.com/releases.csv
- Full dataset as JSON: https://theaiengineering.com/releases.json
