今天,我们发布 QAD Q4_0 GGUF。这些是 LFM2.5-230M、LFM2.5-350M、LFM2.5-1.2B-Instruct 和 LFM2.5-2.6B 的更新版 4-bit checkpoint。它们让开发者能够在 Q4_0 的内存占用和速度下运行 LFM2.5 模型,而不必承受通常的质量下降:
- 采用量化感知蒸馏(QAD)训练:将高精度教师模型蒸馏到量化学生模型中
- 与原生 Q4_0 相同的内存占用和速度:它们保留了 Q4_0 GGUF 的低内存占用和高吞吐量
- 恢复程度:量化所损失的 BF16 平均准确率恢复了 97%
基准测试结果
对于全部四个模型,我们在涵盖推理、指令遵循、工具使用和智能体能力的基准套件上,将采用训练后量化(PTQ)生成的已发布 GGUF 与经过训练的 QAD Q4_0 checkpoint 进行对比:GPQA Diamond、MMLU-Pro、IFEval、IFBench、Multi-IF 和 BFCLv4。
BF16 GGUF 作为同格式下的上限。我们还加入了一项与规模相称的数学评测:LFM2.5-230M 和 LFM2.5-350M 使用 GSM8K,LFM2.5-1.2B-Instruct 和 LFM2.5-2.6B 使用 AIME25。
我们报告五次重复的平均值。
在全部四个模型中,QAD 大幅提升了 Q4_0 checkpoint。QAD checkpoint 分别保留了各自 BF16 基线性能的 97.1%、96.5%、97.4% 和 96.6%。
真实边缘硬件上的速度与体积
我们测量了四款模型 LFM2.5-230M、LFM2.5-350M、LFM2.5-1.2B-Instruct 和 LFM2.5-2.6B 在四个目标平台上的解码吞吐量:MacBook Pro、NucBox EVO-X2、Samsung Galaxy S26 Ultra 和 Raspberry Pi 5。
MacBook Pro 和 NucBox 使用 GPU 推理,而 Samsung 和 Raspberry Pi 使用 Arm CPU 推理。在进行了性能分析的情况下,BF16 和 F16 作为全精度参考一并展示。
230M 和 350M 的 QAD Q4_0 checkpoint 在评估方差范围内达到了与 Q5_K_M 相当的质量,同时解码吞吐量高出 4-33%。1.2B 和 2.6B 的 QAD Q4_0 checkpoint 达到了与 Q4_K_M 相当的质量,吞吐量高出 3-14%。
QAD Q4_0 checkpoint 在适用情况下(即 230M 和 1.2B)也达到了与 Unsloth 的 UD-Q4_K_XL 相当的水平,后者是一个强大的外部训练后量化 checkpoint。
如何使用 QAD GGUF
将这些文件与 llama.cpp 或任何支持 GGUF Q4_0 产物的运行时配合使用。
llama-cli -hf LiquidAI/LFM2.5-350M \
--hf-file LFM2.5-350M-QAD-Q4_0.gguf \
-p "What is C. elegans?"
开始使用 QAD GGUF
QAD GGUF 现已在 Hugging Face 上线:LFM2.5-230M、LFM2.5-350M、LFM2.5-1.2B-Instruct 和 LFM2.5-2.6B。
我们迫不及待想看到你构建的作品。
引用
如需引用,请使用以下参考文献或 BibTeX:
Liquid AI, "LFM2.5 Q4_0: Quantization-Aware Distillation for Edge Deployment", Liquid AI Blog, Aug 2026.
或使用 BibTeX 引用
@article{liquidAI2026Q40,
author = {Liquid AI},
title = {LFM2.5 Q4_0: Quantization-Aware Distillation for Edge Deployment},
journal = {Liquid AI Blog},
year = {2026},
note = {www.liquid.ai/blog/qad},
}
Today, we release QAD Q4_0 GGUFs. These are updated 4-bit checkpoints for LFM2.5-230M, LFM2.5-350M, LFM2.5-1.2B-Instruct, and LFM2.5-2.6B. They allow developers to run LFM2.5 models at Q4_0 memory and speed without the usual quality drop:
- Trained with Quantization-Aware Distillation (QAD): a high-precision teacher model is distilled into a quantized student model
- Same memory and speed as native Q4_0: They keep the low memory footprint and high throughput of Q4_0 GGUFs
- Recovery: 97% of their BF16 average accuracy lost to quantization is recovered
Benchmark results
For all four models, we compare their released GGUFs produced with post-training quantization (PTQ) against the trained QAD Q4_0 checkpoints on a benchmark suite spanning reasoning, instruction-following, tool use, and agentic capabilities: GPQA Diamond, MMLU-Pro, IFEval, IFBench, Multi-IF, and BFCLv4. The BF16 GGUF serves as the in-format ceiling. We also add one scale-appropriate math evaluation: GSM8K for LFM2.5-230M and LFM2.5-350M, and AIME25 for LFM2.5-1.2B-Instruct and LFM2.5-2.6B. We report the mean across five repeats.
Across all four models, QAD substantially improves the Q4_0 checkpoint. The QAD checkpoints retain 97.1%, 96.5%, 97.4%, and 96.6% of their respective BF16 baseline performance.
Speed and size on real edge hardware
We measure decode throughput for the four models LFM2.5-230M, LFM2.5-350M, LFM2.5-1.2B-Instruct, and LFM2.5-2.6B across four targets: MacBook Pro, NucBox EVO-X2, Samsung Galaxy S26 Ultra, and Raspberry Pi 5. MacBook Pro and NucBox use GPU inference, while Samsung and Raspberry Pi use Arm CPU inference. BF16 and F16 are shown as full-precision references where profiled.
The 230M and 350M QAD Q4_0 checkpoints match Q5_K_M quality within evaluation variance at a 4-33% higher decode throughput. The 1.2B and 2.6B QAD Q4_0 checkpoints match Q4_K_M quality at a 3-14% higher throughput. The QAD Q4_0 checkpoints also match Unsloth's UD-Q4_K_XL (where applicable, for the 230M and 1.2B), a strong external post-training quantization checkpoint.
How to use QAD GGUFs
Use the files with llama.cpp or any runtime that supports GGUF Q4_0 artifacts.
llama-cli -hf LiquidAI/LFM2.5-350M \
--hf-file LFM2.5-350M-QAD-Q4_0.gguf \
-p "What is C. elegans?"
Get Started with QAD GGUFs
The QAD GGUFs are available on Hugging Face today: LFM2.5-230M, LFM2.5-350M, LFM2.5-1.2B-Instruct, and LFM2.5-2.6B.
We can't wait to see what you build.
Citation
For citations, please use the following reference or BibTeX:
Liquid AI, "LFM2.5 Q4_0: Quantization-Aware Distillation for Edge Deployment", Liquid AI Blog, Aug 2026.
Or use the BibTeX citation
@article{liquidAI2026Q40,
author = {Liquid AI},
title = {LFM2.5 Q4_0: Quantization-Aware Distillation for Edge Deployment},
journal = {Liquid AI Blog},
year = {2026},
note = {www.liquid.ai/blog/qad},
}




