采用LFM3：专门情报的每一种模式
今天我们很高兴宣布LFM3, LFM3模式家族自然是多模式的，带来文本，视觉，音频，omni，以及决定模型都进入了一个新的共享骨干。 LFM3家族的首次发布涵盖了2B参数范围内的模型。
LFM3建于新的混合式建筑和22T预训令牌上，在保持其效率的同时，比LFM2更具有表现力。 在一个小而快的模型中的额外能力使得LFM3成为其专业智能类中的理想起点：这是因为它可以在当地进行微调，可以在任何设备上运行，并且它足以成为专业任务的可靠专家。
基础和后训练的模型是开放量级的，今天可以在Hugging Face，我们的游乐场，和液态AI API上获得。 看看我们的文件如何运行和微调他们的地方。
下载权重
增加表达力的新架构
LFM2显示，由短演化和注意力组成的混合体可以在不牺牲质量的情况下快速跑出端侧。 它的设计来自于我们的自动硬件即时架构搜索。 对于LFM3，我们利用这一搜索重新设计了用于代理使用案例的架构，这些案例需要每个参数更多的能力，并且需要长时间的快速推论。
我们还将循环中的硬件从一台手机的CPU和一台LFM2的膝上型计算机扩充到14个设备以及LFM3的后端组合：手机(Samsung Galaxy S26 Ultra和OnePlus 15)，一款Raspberry Pi 5，笔记本电脑(Apple M4 Max和AMD Ryzen AI)，和Arm和x86云CPU，外加GPU和Qualcomm Hexagon NAU.
刚开始的短暂进化仍然占了一半的地层为了增加能力而不减慢长期环境，我们尝试将其中一些换成其他顺序块，滑入窗口(SWA)的注意力提供了最佳质量-速度取舍。 我们只将下文所述的其他变化保留在那些没有有意义的速度成本的情况下提高质量的地方。 长上下文取出显示。 每条参数，LFM3-1.7B与LFM2-1.2B在跨边设备上的短提示相匹配，在上下文中与8K个令牌相匹配的速度约为25%.
三合层相接。 LFM3结合了三个操作员：入门短演化，滑出-窗口关注(SWA)，以及全球关注。 在它的24个层中，有13个是演化体，有5个是带有256个托肯窗口的SWA，有6个是全球关注。 演化还使用了由LFM2中的3个起起的7个内核大小。
注意头部一个 sigmoid 门对每个头的输出进行分级，当头没有东西可以添加时就可以关闭。 这有助于抑制破坏量化的活化异常，并增强训练稳定性。
部分 RoPE. 全球关注层每个头只旋转出最高频率的一半并留下低频率半个未旋转，所以无论距离，这些维度都与内容相匹配。 SWA层保持完整的ROPE.
三明治规范。 每个区块都使其输入和输出都正常化，这有助于使活化度保持良好的尺度。
LFM3使用与LFM2.5-8B-A1B相同的128K词汇来表示非拉丁文文字的代名词。
培训。 LFM3-1.7B-Base在22T令牌上预训。 与LFM2相比，混合了新增更多的科学论文，寄存器级代码，更广泛的知识来源，并大大地增加了数学。 最终的3.5T令牌是一个关于被整理，质量更高的数据的中试阶段，具有较大比例的代理痕迹和多种工具调用来为代理后训练准备模型。 这个阶段还将上下文窗口扩展至256K个令牌。
满足各种需要的多模式家庭
第一个LFM3发布包括了6个型号，全部从同一种LFM3-1.7B-Base主干线分出并全部围绕2B参数。 其中四个涵盖不同的输入模式(文字、视觉、音频和omni)，而后两个模式是从视觉和omni模式中专门化的决策模式。 与其运行一个大型通用型号，不如选取一个适合您使用的外壳，自定义的外壳来运行端侧。
为了更深入地潜入这些模型背后的性能和推论结果，请参考相应的模型卡。
文本
LFM3-1.7B是本地agents的文本模型和具有256K-token上下文窗口的专门应用。
在赫尔梅斯Agent，皮和OpenClaw等代理机上运行是经过培训后进行的。 它可以分解您的收件箱，管理您的日历，在网络上研究话题，发送您关于WhatsApp的晨间简报，并使用工具搜索在上千个选项中找到正确的工具。 由于代币是免费的，数十个agents可以在背景中连续运行。 您可以根据您的域名、 绳索和工具来调整它，在一个单一的GPU上运行一个短的微调。
在代理任务上，LFM3-1.7B与型号相竞争，其大小是其数倍。 它在引导指令跟随和功能上调用 BFCL 和 ToolSandbox，并在浏览器和商务app(WebArena-Verifed，自动化奔驰)的长任务上击败了Qwen3.5-9B. 更大的模型仍然在代理编码和知识上取胜，其中大小最重要，但搜索有助于缩小知识差距，几乎是人类最后考试的分数的两倍。
为了取得这些成果，该模式经历了四个培训后阶段：
- 监督微调(SFT):1.75 T符号跨越了各种任务，包括聊天，推理，工具调用，以及代理痕迹。
- 混合强化学习：从数学和代码到SQL和结构化产出，同时培训许多领域。 奖励中的长罚使答案简明。
- 多领域政策提炼(MOPD):6名专业RL教师(代码、数学、工具使用、长背景、代理工作和企业工作流程)被提炼成单一模式。
- 代理RL：从网页浏览到办公室工作的任务都在真正的agent带内运行，为部署模型做准备。
端侧，我们用Q4 0的llama.cpp和一个32K-token上下文来测量度量。 LFM3-1.7B是所有三种设备上最快的模型，解码比下一种快1.8到2.8倍，它使用最少的内存。 长上下文仍然便宜：完整的128K-token上下文需要1.6GB的KV缓存，而Qwen3.5-4B需要4.3GB,MiniCPM5-2B需要5.6GB.
在一单NVIDIA B300上用BF16的SGLang,LFM3-1.7B每天能以1,024个同时请求而出出出36亿个令牌，比这个比较中的任何其他模型都要多。
为了显示这种吞吐量使得什么成为可能，我们同时运行了30多万个LFM3-1.7B agents，以调查安然电子邮件档案。
LFM3-1.7B可在随机流量下以低潜速提供，使实时在Liquid RIPL上运行时使用在线欺诈检测等案例。 P95请求的耐久性仍然稳定，即使负重，在vLLM上提供的相竞模型也表现优异。
愿景
LFM3-VL-2B可以观看，理性并理解物理世界：
- 视频理解：LFM视觉模型首次将视频作为输入。 询问有关录制的讲座、你手机上拍摄的走道或相机录像的片段的问题。
- 思考：模型的原因在回答之前一步一步地出现。 这有助于它更好地理解用户输入，并解决更难的视觉问题，如数学，图表，以及多步骤的问题。
- 空间和内含的推理：模型将物体定位的能力及其空间关系的理由为下游机器人应用奠定了坚实的基础。
LFM3-VL-2B是比起LFM2.5-VL系列的性能飞跃。 在62个视星等的基准测试上，LFM3-VL-2B至少以7.4分的分数大大地领先了所有大小相近的型号，尽管其分数很大，它还是与Gemma 4 12B的性能相匹配，一个型号大了6倍多，超越 Gemini 3.5 Flash Lite，并位于GPT-5.6 Luna的2分以内。 更详细情况请参见"模范卡".
指出：每个类别分数是其基准测试分数的未加权平均值。 平均行数是所有62项评价中未加权的平均值。
这一切可以高效地运行端侧，它使得日常使用智能而无需数据离开你的设备或需要云连接。 端侧，LFM3-VL-2B在所有被测量设备中具有最高的预填吞吐量和最低的内存足迹，相较于其大小级的其他模型。
在BF16装有SGLang的单个NVIDIA B300上，LFM3-VL-2B每秒能出出出超过一万二千个令牌，高于其他同大小级的型号。
LFM3-VL-2B可以在随机流量下以低空提供，使实时在Liquid RIPL上运行时使用像图像资产标记等案例。 P95请求的耐久性仍然稳定，即使负重，在vLLM上提供的相竞模型也表现优异。
视觉编码器
除了LFM3主干线的改进外，LFM3-VL还伴随着我们新的内部视觉编码器，该编码器在之前的LFM VL模型中使用了比SigLIP2编码器少的3×.
液态视觉编码器结合了进取和注意力，使其在同等质量上明显地比ViT对应器更有效。 下图显示这一点。 我们的164M-参数编码器与Qwen3.5-2B的视觉编码器相匹配，平均精度为55.9%，跨越了13个视觉语言模型基准测试，同时需要减少2.2×FLOP.
对于视频，编码器会将每对被采样的相框共同嵌入到一对时间片的视觉特征上，将时间序列长度减半后再进一步编码。 这些选择共同使编码器成为更丰富图像理解和高效视频处理的基础。
音频
LFM3-Audio-2B是一个紧凑，功能齐全的音频模型，在笔记本电脑和边缘设备上在当地运行。 它以文本或音频作为输入，并以文本，语音，或两者兼而有之来响应。
LFM3-Audio-2B采用了新的架构，内部开发了音频编码器和编码器。 它在语音质量上超过了LFM2.5-Audio，现在支持18种语言。 新的能力包括音频理解，通用音频推理，以及工具调用。
- 语音聊天：大声提问，在文字和讲话中收到流传的答复。 对于只使用文字的回复，模型在回答之前也可以说明理由。
- 多语种语音：翻译、合成和翻译18种语言的语音。
- 音频理解和推理：请模型解释录音并回答关于演讲、声音和包含的事件的问题。
- 工具呼叫：将口头请求变成控制应用程序和设备的工具呼叫。 该模型还可以在发出这些呼吁的同时以文字或发言方式做出回应。
ASR (WER,%，下级更好)
英语 TTS (UTMOSv2, 1-5，较高更好) (WER,%，较低更好)
声音奔驰，工具呼叫，音频理解(音频输入，文本输出，%，更高更好)
我们新的基于LFM的音频编码器是利用滑入窗口的注意力和短演层混合而成，保持内存利用率的界限。 新编码器是完全因果的，允许模型在说话时处理活的麦克风输入并进行预填。 这会减少你完成后的等待，时间到第一个符号低至50毫秒。
液态Codec，我们的新音频代号，在LFM3-Audio和LFM3-Omni中都具有多语种语音生成能力。 由LFM3架构所建，LiquidCodec以单CPU线条(Apple M4 Max)上约87×实时将生成的活字变为语音，支持流码解码。
量子化将每个80 ms的语音压缩成12个离散的音符，与之前的LFM音频模型形成对比，允许多个音符进行迭代预测，而不是需要相继预测。 由于骨干产生出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出入出出出出出出出出出出出出出出出出出出出出出出出出出入出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出入出出出出出出出出出出出出出出出出出出出出出出出出出出入出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出出
欧尼
LFM3-Omni-2B是我们第一个全能模型，将视觉，音频，语言统一在一个单一的模型中，能够看见，倾听，回声：
- 说到视觉问题
本文由 AI宝库自动抓取并翻译，版权归原作者所有；如需查看原文请点下方链接。
Introducing LFM3: Every Modality for Specialized Intelligence
Today, we’re excited to announce LFM3, our new generation of Liquid Foundation Models (LFMs) built for specialized intelligence. The LFM3 model family is multimodal by nature and brings text, vision, audio, omni, and decision models onto one new shared backbone. The first release of the LFM3 family covers models in the 2B parameter range.
Built on a new hybrid architecture and 22T tokens of pre-training, LFM3 is far more expressive than LFM2 while maintaining its efficiency. The extra capacity in a small, fast model makes LFM3 the ideal starting point in its class for specialized intelligence: this is because it can be fine-tuned locally, it can run on any device, and it is capable enough to become a reliable expert on a specialized task.
The base and post-trained models are open-weight and available today on Hugging Face, our Playground, and the Liquid AI API. Check out our docs on how to run and fine-tune them locally.
Download weightsTry in Liquid APITry in PlaygroundBenchmark in Pipette
New architecture for more expressivity
LFM2 showed that a hybrid of short convolutions and attention could run fast on-device without sacrificing quality. Its design came from our automated hardware-in-the-loop architecture search. For LFM3, we used this search to redesign that architecture for agentic use cases, which need more capability per parameter and fast inference over long contexts.
We also expanded the hardware in the loop, from CPUs in one phone and one laptop for LFM2 to 14 device and backend combinations for LFM3: phones (Samsung Galaxy S26 Ultra and OnePlus 15), a Raspberry Pi 5, laptops (Apple M4 Max and AMD Ryzen AI), and Arm and x86 cloud CPUs, plus GPUs and the Qualcomm Hexagon NPU.
Gated short convolutions still make up over half the layers. To add capability without slowing down long contexts, we tried swapping some of them for other sequence blocks, and sliding-window attention (SWA) gave the best quality–speed tradeoff. We kept the other changes described below only where they improved quality without a meaningful speed cost. The payoff shows at long context. Per parameter, LFM3-1.7B matches LFM2-1.2B on short prompts across edge devices, and with 8K tokens in context it is about 25% faster.
Tri-hybrid layer mix. LFM3 combines three operators: gated short convolutions, sliding-window attention (SWA), and global attention. Of its 24 layers, 13 are convolutions, 5 are SWA with a 256-token window, and 6 are global attention. The convolutions also use a kernel size of 7, up from 3 in LFM2.
Head-wise gated attention. A sigmoid gate scales each head’s output, so a head can switch off when it has nothing to add. This can help suppress activation outliers that hurt quantization and improve training stability.
Partial RoPE. Global attention layers rotate only the highest-frequency half of each head and leave the low-frequency half unrotated, so those dimensions match on content regardless of distance. SWA layers keep full RoPE.
Sandwich norm. Each block normalizes both its input and its output, which helps keep activations well-scaled.
LFM3 uses the same 128K vocabulary as LFM2.5-8B-A1B, built to tokenize non-Latin scripts more efficiently.
Training. LFM3-1.7B-Base was pre-trained on 22T tokens. Compared with LFM2, the mix adds more scientific papers, repository-level code, a wider range of knowledge sources, and considerably more math. The final 3.5T tokens are a mid-training stage on curated, higher-quality data, with a larger share of agentic traces and diverse tool calls to prepare the model for agentic post-training. This stage also extends the context window to 256K tokens.
A multimodal family for every need
The first LFM3 release includes six models, all branched from the same LFM3-1.7B-Base backbone and all around 2B parameters. Four of them cover different input modalities (text, vision, audio, and omni), and the last two are decision models specialized from the vision and omni models. Instead of running one large general-purpose model, you can pick the one that fits your use case, customize it, and run it on-device.
For a deeper dive into the performance and inference results behind these models, refer to the respective model cards.
Text
LFM3-1.7B is a text model for local agents and specialized applications with a 256K-token context window.
It is post-trained to run in agentic harnesses such as Hermes Agent, Pi, and OpenClaw. It can triage your inbox, manage your calendar, research topics on the web, send you morning briefings on WhatsApp, and use tool search to find the right tool among thousands of options. Since tokens are free, dozens of agents can run continuously in the background. You can adapt it to your domains, harnesses, and tools with a short fine-tuning run on a single GPU.
On agentic tasks, LFM3-1.7B competes with models several times its size. It leads on instruction following and function calling on BFCL and ToolSandbox, and beats Qwen3.5-9B on long tasks in browsers and business apps (WebArena-Verified, AutomationBench). Larger models still win on agentic coding and knowledge, where size matters most, but search helps close the knowledge gap, nearly doubling its score on Humanity's Last Exam.
To achieve these results, the model went through four stages of post-training:
- Supervised fine-tuning (SFT): 1.75T tokens across various tasks, including chat, reasoning, tool calling, and agentic traces.
- Mixed reinforcement learning: many domains trained at once, from math and code to SQL and structured outputs. A length penalty in the reward keeps answers concise.
- Multi-domain on-policy distillation (MOPD): six specialist RL teachers (code, math, tool use, long context, agentic work, and enterprise workflows) are distilled back into a single model.
- Agentic RL: tasks from web browsing to office work are run inside real agent harnesses, preparing the model for deployment.
On-device, we measure metrics with llama.cpp at Q4_0 and a 32K-token context. LFM3-1.7B is the fastest model on all three devices, decoding 1.8 to 2.8 times faster than the next one, and it uses the least memory. Long contexts stay cheap: a full 128K-token context needs 1.6 GB of KV cache, against 4.3 GB for Qwen3.5-4B and 5.6 GB for MiniCPM5-2B.
On a single NVIDIA B300 with SGLang in BF16, LFM3-1.7B can produce 3.6 billion tokens per day at 1,024 concurrent requests, more than any other model in this comparison.
To show what this throughput makes possible, we ran more than 300,000 LFM3-1.7B agents in parallel to investigate the Enron email archive.
LFM3-1.7B can be served at low latency under random traffic, enabling real-time use cases like online fraud detection when run on Liquid RIPL. P95 request latency remains stable even under heavy load, outperforming competing models served on vLLM.
Vision
LFM3-VL-2B can watch, reason, and understand the physical world:
- Video understanding: for the first time, an LFM vision model takes video as input. Ask questions about a recorded lecture, a walkthrough filmed on your phone, or a clip from a camera feed.
- Thinking: the model reasons step by step before answering. This helps it better understand user inputs and solve harder visual questions such as math, charts, and multi-step problems.
- Spatial and embodied reasoning: the model’s ability to localize objects and reason about their spatial relationships provides a strong foundation for downstream robotics applications.
LFM3-VL-2B is a leap in performance over the LFM2.5-VL series. Across 62 vision benchmarks, LFM3-VL-2B scores significantly ahead of all models of similar size by at least 7.4 points, and despite its size, it matches the performance of Gemma 4 12B, a model more than 6 times larger, surpasses Gemini 3.5 Flash Lite and is within 2 points of GPT-5.6 Luna. For more details, see the model card.
Notes: Each category score is the unweighted mean of its benchmark scores. The Average row is the unweighted mean across all 62 evaluations.
All of this can be run efficiently on-device, which enables everyday use of intelligence without the data leaving your device or the need for a cloud connection. On-device, LFM3-VL-2B has the highest prefill throughput and lowest memory footprint across all measured devices, compared to other models of its size class.
On a single NVIDIA B300 with SGLang in BF16, LFM3-VL-2B can produce over 12,000 tokens per second, higher than any other model of the same size class.
LFM3-VL-2B can be served at low latency under random traffic, enabling real-time use cases like image asset tagging when run on Liquid RIPL. P95 request latency remains stable even under heavy load, outperforming competing models served on vLLM.
Vision encoder
In addition to improvements to the LFM3 backbone, LFM3-VL comes with our new in-house vision encoder which uses 3× less compute than the SigLIP2 encoder in previous LFM VL models.
The Liquid vision encoder combines convolution and attention to make it significantly more efficient than its ViT counterpart at the same quality. The performance-efficiency graph below illustrates this. Our 164M-parameter encoder matches Qwen3.5-2B’s vision encoder at 55.9% average accuracy across 13 vision-language model benchmarks, while requiring 2.2× fewer FLOPs.
For video, the encoder jointly embeds each pair of sampled frames into one temporal slice of visual features, halving the temporal sequence length before further encoding. Together, these choices make the encoder a foundation for both richer image understanding and efficient video processing.
Audio
LFM3-Audio-2B is a compact, fully featured audio model that runs locally on laptops and edge devices. It takes text or audio as input and responds with text, speech, or both.
LFM3-Audio-2B uses a new architecture with an audio encoder and codec developed in-house. It outperforms LFM2.5-Audio in speech quality and now supports 18 languages. New capabilities include audio understanding, general-purpose audio reasoning, and tool calling.
- Voice chat: ask questions aloud and receive streaming replies in text and speech. For text-only responses, the model can also reason before answering.
- Multilingual speech: transcribe, synthesize, and translate speech across 18 languages.
- Audio understanding and reasoning: ask the model to interpret recordings and answer questions about the speech, sounds, and events they contain.
- Tool calling: turn spoken requests into tool calls to control apps and devices. The model can also respond in text or speech alongside those calls.
ASR (WER, %, lower is better)
English TTS (UTMOSv2, 1-5, higher is better) (WER, %, lower is better)
VoiceBench, Tool calling, Audio understanding (audio in, text out, %, higher is better)
Our new LFM-based audio encoder is built from a mixture of sliding-window attention and short-convolution layers, keeping memory utilization bounded. The new encoder is fully causal, allowing the model to process live microphone input and perform prefill while you speak. This reduces the wait after you finish, with time to the first token as low as 50 ms.
LiquidCodec, our new audio tokenizer, powers multilingual speech generation in both LFM3-Audio and LFM3-Omni. Built on the LFM3 architecture, LiquidCodec turns generated tokens into speech at about 87× real time on a single CPU thread (Apple M4 Max), supporting streaming decoding.
Quantization compresses each 80 ms of speech into 12 discrete tokens and, in contrast to previous LFM Audio models, allows multiple tokens to be predicted iteratively in parallel rather than requiring sequential prediction. As the backbone generates new tokens, the detokenizer progressively converts them into speech, all faster than real-time so that conversations are smooth and responsive.
Omni
LFM3-Omni-2B is our first omni model, bringing vision, audio, and language together in a single model that can see, listen, and talk back:
- Spoken visual question answ
Auto-fetched and translated by AI Gem Box. All rights belong to the original author.