LiquidAI 发布面向端侧的 d1-3B 与 d1-omni-600M 开源决策模型
Multimodal open d1 decision models for the edge
LiquidAI 发布基于 LFM 的两款开源决策模型 d1-3B 和 d1-omni-600M，它们不生成 token，而是在单次前向传播中给出答案。d1-3B 基于 LFM2.5-VL-3B 训练，接受文本和图像输入，在七个公开数据集上平均得分 82.9，高于 Decider 4B；d1-omni-600M 基于双向编码器 LFM2.5-Encoder-350M，加入视觉和音频编码器，得分 78.4，超过 Decider 2B 的 77.1，参数量仅为其四分之一，目前处于早期研究发布阶段。与 NVIDIA 合作在 RTX 4090、Jetson AGX Thor、Jetson AGX Orin 64 GB 和 Jetson Orin Nano 上评测，d1-3B 在每台设备上回答单个问题均低于 50 ms，GPU 上低于 10 ms，处理 384px 图像低于 18 ms。两款模型均为开放权重，已在 Hugging Face 上线。
推荐理由
AI 评分 55
LiquidAI 公开两款决策模型在端侧与 GPU 上的延迟数据，可据此判断单次前向决策在边缘场景的可用性。
主题
正文
当前显示原文，中文全文译文尚未就绪。
How we built decision models for the edge
These open d1 decision models are built on our Liquid Foundation Models (LFMs). Unlike our generative models, decision models don’t produce tokens but answer in a single forward pass.
d1-3B and d1-omni-600M are trained from two very different backbones:
d1-3B is trained from LFM2.5-VL-3B , our latest VLM, which is decoder-only. It accepts text and images as inputs.
d1-omni-600M is trained from LFM2.5-Encoder-350M , a bidirectional encoder. It adds vision and audio encoders to handle all three modalities. It accepts either text and image, or text and audio as inputs. This model is currently in an early research release and is undergoing further development.
Benchmark results
We benchmarked d1-3B and d1-omni-600M on seven public datasets spanning reading comprehension, toxicity detection, intent classification, medical QA, and cross-lingual understanding. d1-3B achieves a mean score of 82.9, the highest in the table and above Decider 4B. d1-omni-600M scores 78.4, surpassing Decider 2B (77.1) with only a quarter of the parameters.
We validated that d1-3B retains the vision capabilities of its LFM2.5-VL-3B backbone on standard vision benchmarks, and that d1-omni-600M handles all three modalities. We do not report any vision or audio benchmarks, as the Decision Index v0.3 includes only a private vision split and audio decision benchmarks are currently an open problem.
Speed
In collaboration with NVIDIA, we evaluated d1-3B on the NVIDIA stack across NVIDIA GeForce RTX 4090, NVIDIA Jetson AGX Thor, Jetson AGX Orin 64 GB, and Jetson Orin Nano. Since d1-omni-600M is an early research release, we don’t report any speed numbers for it in this release.
Edge inference. d1-3B answers a single question in under 50 ms on every measured device. Three questions take only 1.3x the time of one, with the AGX Thor going from 16 ms to 20 ms.
GPU inference. On GPU, d1-3B answers a question in under 10 ms and processes a 384px image in under 18 ms on both platforms.
How to use open d1 decision models
Reach for d1 decision models when you need fast, structured decisions, including multimodal inputs. d1-3B delivers the highest decision quality at its size, while d1-omni-600M fits where footprint matters.
Install the dependencies (requires transformers>=5.14 ):
These model ship their own code, so load it with trust_remote_code=True :
For brevity, we only include the example for d1-3B. See the d1-omni-600M model card for instructions on how to run it.
Both decision models are open-weight and available on Hugging Face today:
Download: d1-3B and d1-omni-600M on Hugging Face.
Try: run the demos in our System One Arcade Hugging Face Space.
We can't wait to see what you build.
Citation
If you use this work, please cite the release blog:
Models mentioned in this article 2
LiquidAI/LFM2.5-Encoder-350M
LiquidAI/LFM2.5-VL-3B
More from this author
Accelerating vision-language models with LFM2.5-VL-DSpark
Up to 3.2x Faster Inference with LFM2.5-DSpark
Models mentioned in this article 2
LiquidAI/LFM2.5-Encoder-350M
LiquidAI/LFM2.5-VL-3B