液态AI 发布 LFM3 6个开放-重量2B级模型家族

AI 模型和平台

液态AI 发布 LFM3 6个开放-重量2B级模型家族

在 Google 上的首选来源中添加 Unite.AI

液态AI 宣布 LFM3,2026年10月27日，新一代开放量级液态基础模型，其带来文本，视觉，音频，omni，以及决定模型进入一个共享骨干。 Hugging Face博客的一篇文章详细介绍了首个发布，涵盖2B参数范围内的六个模型。

共享背骨上的六个模型

每个LFM3型号的分支来自同一个LFM3-1.7B-Base. LFM3-1.7B 接收文本并返回文本。 LFM3-VL-2B接受文本，图像，视频并返回文本，而LFM3-Audio-2B接收文本或音频，并用文本或语音响应。 LFM3-Omni-2B,Liquid AI将其描述为第一个omni模型，接受文本，图像，视频，音频，并产生文本和语音。 两个决定模型，即d1-2B和d1-omni-2B，都是从视觉模型和omni模型以及返回校准的答案而不是生成文本中专门化的。

根据LFM3-1.7B型卡，文本模型有1.73B参数，上下文长度为256K-token，上下文词汇长度为12.8万，支持16种语言，在LFM Open License v1.0下发船。 基础型和经培训后型号为开放重量型，可在Hugging Face，液态AI游戏场，液态AIAPI上获得。

液态AI的发布帖名日-0发布合作伙伴Ambarella,AMD,Arm,Intel通过OpenVINO,NVIDIA,OpenRouter,Qualcomm通过QualcommAI Hub和GenieX,Vercel，称奔驰和Shopify合作进行发布前测试和评价。

三合院建筑和培训

LFM3-1.7B-Base共有24层：13个门质短旋，5个有256个托肯窗口的滑窗关注层，以及6个全球关注层。 演化体使用内核大小为7，比LFM2中的3高。 还有三处变化定义了主干线：头部有门的注意，其中，一个Sigmoid门对每个注意头的输出进行测平，并可以压制伤害四分位化的激活输出；部分RoPE，其中全球注意层只旋转每个头的最高频率的一半，而滑动的窗口层则保持完整的RoPE；以及一个三明治规范，使每个块的输入和输出正常化。

设计来自Liquid AI的硬件内置架构搜索，扩展为LFM3的14个设备和后端组合，包括手机(Samsung Galaxy S26 Ultra和OnePlus 15)，一台Raspberry Pi 5，笔记本电脑(Apple M4 Max和AMD Ryzen AI),Arm和x86云CPU,GPU，以及Qualcomm Hexagon NPU.

LFM3-1.7B-Base在22T令牌上接受了预训，最终的3.5T令牌在曲解数据上形成了一个中训阶段，其中具有较大比例的代理痕迹和工具呼叫，这也将上下文窗口扩展至256K令牌。 培训后分四个阶段进行：在1.75T令牌上监督微调，混合强化学习，6名专业强化学习教师的多领域政策提炼，以及真agent带内的代理强化学习。

液态AI 称新架构允许LFM3-1.7B以35%的参数超过其前身LFM2.5-2.6B，每个参数在短提示上与LFM2-1.2B匹配，同时在上下文中与8K令牌运行速度快约25%. 在4位精度上，公司称，文本模型需要约1GB的内存。

公司报告的业绩和专业

Liquid AI报告说，在WebArena-Verifed上LFM3-1.7B的分数为38.42，而LFM2.5-2.6B的分数为3.02，在代理任务上收益最大，尽管该型卡称该型号因其体积小，不推荐用于代理编码或复杂的多步推理。

液态AI 称 LFM3-VL-2B是第一个接受视频输入的LFM视觉模型，并报告说它融资是62视觉的平均值基准测试，从57.6到71.4，与LFM2.5-VL-3B相比，同时对3.1B使用1.9B参数，使其与Gemma 4 12B相当，这个模型大六倍以上。 一个新的内部164M参数的视觉编码器使用比之前的LFM VL模型中的SigLIP2编码器少的3x计算，并且根据公司的测量结果将Qwen3.5-2B的视觉编码器的精度与更少的2.2xFLOP匹配。

LFM3-Audio-2B支持18种语言，Liquid AI报告在UTMOSv2上的语音质量为3.83，而LFM2.5-Audio-1.5B的语音质量为3.22，这是一个完全因果的音频编码器，时间最短至50 ms，以及一个LiquidCodec的代碼器，在Apple M4 Max的单个CPU线条上以约87x实时生成语音。 公司称 2.2B参数LFM3-Omni-2B击败了包括Qwen2.5-Omni-7B在内的10.7B参数的型号，在Daily-Omni,WorldSense，以及AV-Speaker Bench视听理解基准测试上。

决策模型在单个前传中回答一系列问题，每个问题都有校准的概率，并在对象处添加键入的答案，从文档中读取值，以一个范围估计数字，并选择每一个应用的选项。 液态AI 称 d1-2B将d1-3B替换为三分之一的参数，在Imajev Bench上得分75.5比64.0.

每个LFM3型号都可以在单一的GPU上进行微调，Liquid AI出版的笔记本覆盖了被监督的微调，DPO,GRPO，以及对文本，视觉，音频，和omni型号的持续预训。 在一家公司的例子中，LFM3-VL-2B在一台A100上对大约3000张公共风涡扇叶片照片进行了微调，持续数小时，将缺陷分类精度提高到67%，在被搁置的照片上检测到F1从0.003到0.11.

Liquid AI还将LFM3-VL-2B改装成三种机器人模型，用于管弦，导航和操纵。 该公司报告说，操控视觉语言-行动模式在15个模具中进行了培训，在4个LIBERO套房中取得了96.2%的平均成功，在RoboCasa-GR1桌面任务上取得了45.7%的平均成功，同时在TensorRT FP16的Jetson AGX Orin上产生了50ms以下的动作块。

在吞吐量上，液态AI 称 LFM3-1.7B可以在单NVIDIA B300上每天生产36亿令牌，满足1,024个同时请求。 原产地支持llama.cpp,MLX,VLLM的型号飞船横跨Apple,AMD,NVIDIA和Qualcomm硬件，其检查站变体包括BF16,FP8,NVFP4,GGUF,MLX，和ONNX.

本文由 AI图鉴自动抓取并翻译，版权归原作者所有；如需查看原文请点下方链接。

AI Models & Platforms

Liquid AI Launches LFM3 Family of Six Open-Weight 2B-Class Models

Add Unite.AI to your preferred sources on Google

Liquid AI announced LFM3 on October 27, 2026, a new generation of open-weight Liquid Foundation Models that brings text, vision, audio, omni, and decision models onto one shared backbone. The first release, detailed in a post on the Hugging Face blog, covers six models in the 2B parameter range.

Six Models on One Shared Backbone

Every LFM3 model branches from the same LFM3-1.7B-Base. LFM3-1.7B takes text in and returns text. LFM3-VL-2B accepts text, images, and video and returns text, while LFM3-Audio-2B takes text or audio and responds with text or speech. LFM3-Omni-2B, which Liquid AI describes as its first omni model, accepts text, images, video, and audio and produces text and speech. Two decision models, d1-2B and d1-omni-2B, are specialized from the vision and omni models and return calibrated answers rather than generated text.

According to the LFM3-1.7B model card, the text model has 1.73B parameters, a 256K-token context length, a 128,000-token vocabulary, and support for 16 languages, and it ships under the LFM Open License v1.0. The base and post-trained models are open-weight and available on Hugging Face, the Liquid AI Playground, and the Liquid AI API.

Liquid AI’s release post names day-0 launch partners Ambarella, AMD, Arm, Intel via OpenVINO, NVIDIA, OpenRouter, Qualcomm via Qualcomm AI Hub and GenieX, and Vercel, and says Mercedes-Benz and Shopify partnered on pre-release testing and evaluation.

Tri-Hybrid Architecture and Training

LFM3-1.7B-Base has 24 layers: 13 gated short convolutions, five sliding-window attention layers with a 256-token window, and six global attention layers. The convolutions use a kernel size of 7, up from 3 in LFM2. Three further changes define the backbone: head-wise gated attention, in which a sigmoid gate scales each attention head’s output and can suppress activation outliers that hurt quantization; partial RoPE, under which global attention layers rotate only the highest-frequency half of each head while sliding-window layers keep full RoPE; and a sandwich norm that normalizes each block’s input and output.

The design came from Liquid AI’s hardware-in-the-loop architecture search, expanded for LFM3 to 14 device and backend combinations, including phones (Samsung Galaxy S26 Ultra and OnePlus 15), a Raspberry Pi 5, laptops (Apple M4 Max and AMD Ryzen AI), Arm and x86 cloud CPUs, GPUs, and the Qualcomm Hexagon NPU.

LFM3-1.7B-Base was pre-trained on 22T tokens, with the final 3.5T tokens forming a mid-training stage on curated data with a larger share of agentic traces and tool calls that also extended the context window to 256K tokens. Post-training ran in four stages: supervised fine-tuning on 1.75T tokens, mixed reinforcement learning, multi-domain on-policy distillation from six specialist reinforcement-learning teachers, and agentic reinforcement learning inside real agent harnesses.

Liquid AI said the new architecture lets LFM3-1.7B outperform its predecessor LFM2.5-2.6B with 35% fewer parameters, and that per parameter it matches LFM2-1.2B on short prompts while running about 25% faster with 8K tokens in context. At 4-bit precision, the company said, the text model takes about 1 GB of memory.

Company-Reported Performance and Specialization

Liquid AI reports that LFM3-1.7B scores 38.42 on WebArena-Verified against 3.02 for LFM2.5-2.6B, with the largest gains on agentic tasks, though the model card states the model is not recommended for agentic coding or complex multi-step reasoning because of its small size.

Liquid AI said LFM3-VL-2B is the first LFM vision model to accept video input, and reports that it raises the average across 62 vision benchmarks from 57.6 to 71.4 compared with LFM2.5-VL-3B while using 1.9B parameters against 3.1B, putting it on par with Gemma 4 12B, a model more than six times larger. A new in-house 164M-parameter vision encoder uses 3x less compute than the SigLIP2 encoder in previous LFM VL models and matches the accuracy of Qwen3.5-2B’s vision encoder with 2.2x fewer FLOPs, according to the company’s measurements.

LFM3-Audio-2B supports 18 languages, and Liquid AI reports speech quality of 3.83 on UTMOSv2 against 3.22 for LFM2.5-Audio-1.5B, a fully causal audio encoder with time to first token as low as 50 ms, and a LiquidCodec tokenizer that generates speech at about 87x real time on a single CPU thread on an Apple M4 Max. The company said the 2.2B-parameter LFM3-Omni-2B beats models up to 10.7B parameters, including Qwen2.5-Omni-7B, on the Daily-Omni, WorldSense, and AV-SpeakerBench audio-visual understanding benchmarks.

The decision models answer a list of questions in a single forward pass, each with a calibrated probability, and add typed answers that point at objects, read values from documents, estimate numbers with a range, and pick every option that applies. Liquid AI said d1-2B replaces d1-3B with a third fewer parameters and scores 75.5 against 64.0 on ImajevBench.

Every LFM3 model can be fine-tuned on a single GPU, and Liquid AI published notebooks covering supervised fine-tuning, DPO, GRPO, and continued pre-training for the text, vision, audio, and omni models. In one company example, LFM3-VL-2B was fine-tuned on about 3,000 public wind-turbine blade photos for a few hours on one A100, raising defect classification accuracy to 67% and detection F1 from 0.003 to 0.11 on held-out photos.

Liquid AI also adapted LFM3-VL-2B into three robotics models for orchestration, navigation, and manipulation. The company reports that the manipulation vision-language-action model, trained across 15 embodiments, achieved 96.2% average success across the four LIBERO suites and 45.7% on RoboCasa-GR1 tabletop tasks while generating action chunks in under 50 ms on a Jetson AGX Orin with TensorRT FP16.

On throughput, Liquid AI said LFM3-1.7B can produce 3.6 billion tokens per day on a single NVIDIA B300 at 1,024 concurrent requests. The models ship with native support for llama.cpp, MLX, and vLLM across Apple, AMD, NVIDIA, and Qualcomm hardware, with checkpoint variants including BF16, FP8, NVFP4, GGUF, MLX, and ONNX.

Auto-fetched and translated by AI Atlas. All rights belong to the original author.