# FGMP: Fine-Grained Mixed-Precision Weight and Activation Quantization for Hardware-Accelerated LLM Inference

arXiv: https://arxiv.org/abs/2504.14152 · HTML: https://mlsanity.com/paper/2504.14152/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 31 references that are in the map.
- First posted on arXiv 19 Apr 2025.

## Builds on (most cited first, 19 of 19)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers](https://mlsanity.com/paper/2210.17323/) (2022)
- [Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference](https://mlsanity.com/paper/1712.05877/) (2017)
- [AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration](https://mlsanity.com/paper/2306.00978/) (2023)
- [LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scale](https://mlsanity.com/paper/2208.07339/) (2022)
- [QuaRot: Outlier-Free 4-Bit Inference in Rotated LLMs](https://mlsanity.com/paper/2404.00456/) (2024)
- [A White Paper on Neural Network Quantization](https://mlsanity.com/paper/2106.08295/) (2021)
- [SpinQuant: LLM quantization with learned rotations](https://mlsanity.com/paper/2405.16406/) (2024)
- [HAQ: Hardware-Aware Automated Quantization with Mixed Precision](https://mlsanity.com/paper/1811.08886/) (2018)
- [HAWQ: Hessian AWare Quantization of Neural Networks with Mixed-Precision](https://mlsanity.com/paper/1905.03696/) (2019)
- [Atom: Low-bit Quantization for Efficient and Accurate LLM Serving](https://mlsanity.com/paper/2310.19102/) (2023)
- [HAWQ-V2: Hessian Aware trace-Weighted Quantization of Neural Networks](https://mlsanity.com/paper/1911.03852/) (2019)
- [Microscaling Data Formats for Deep Learning](https://mlsanity.com/paper/2310.10537/) (2023)
- [Mixed Precision Quantization of ConvNets via Differentiable Neural Architecture Search](https://mlsanity.com/paper/1812.00090/) (2018)
- [GOBO: Quantizing Attention-Based NLP Models for Low Latency and Energy Efficient Inference](https://mlsanity.com/paper/2005.03842/) (2020)
- [OliVe: Accelerating Large Language Models via Hardware-friendly Outlier-Victim Pair Quantization](https://mlsanity.com/paper/2304.07493/) (2023)
- [VS-Quant: Per-vector Scaled Quantization for Accurate Low-Precision Neural Network Inference](https://mlsanity.com/paper/2102.04503/) (2021)
- [MicroScopiQ: Accelerating Foundational Models through Outlier-Aware Microscaling Quantization](https://mlsanity.com/paper/2411.05282/) (2024)
- [FIT: A Metric for Model Sensitivity](https://mlsanity.com/paper/2210.08502/) (2022)

## Built on by (most cited first, 9 of 9)

- [ARCQuant: Boosting NVFP4 Quantization with Augmented Residual Channels for LLMs](https://mlsanity.com/paper/2601.07475/) (2026)
- [MicroMix: Efficient Mixed-Precision Quantization with Microscaling Formats for Large Language Models](https://mlsanity.com/paper/2508.02343/) (2025)
- [ScaleBITS: Scalable Bitwidth Search for Hardware-Aligned Mixed-Precision LLMs](https://mlsanity.com/paper/2602.17698/) (2026)
- [SNIP: An Adaptive Mixed Precision Framework for Subbyte Large Language Model Training](https://mlsanity.com/paper/2602.01410/) (2026)
- [SFMP: Fine-Grained, Hardware-Friendly and Search-Free Mixed-Precision Quantization for Large Language Models](https://mlsanity.com/paper/2602.01027/) (2026)
- [MoBiQuant: Mixture-of-Bits Quantization for Token-Adaptive Any-Precision LLM](https://mlsanity.com/paper/2602.20191/) (2026)
- [SharQ: Bridging Activation Sparsity and FP4 Quantization for LLM Inference](https://mlsanity.com/paper/2606.26587/) (2026)
- [Saturation Makes Quantization Error Additive: A Coverage Model with a Certificate](https://mlsanity.com/paper/2607.12266/) (2026)
- [AlignQuant: Tile-Aligned Mixed-Precision Quantization for Efficient LLM Generation](https://mlsanity.com/paper/2610.07457/) (2026)

## Abstract (arXiv)

Quantization is a powerful tool to improve large language model (LLM) inference efficiency by utilizing more energy-efficient low-precision datapaths and reducing memory footprint. However, accurately quantizing LLM weights and activations to low precision is challenging without degrading model accuracy. We propose fine-grained mixed precision (FGMP) quantization, a post-training mixed-precision quantization hardware-software co-design methodology that maintains accuracy while quantizing the majority of weights and activations to reduced precision. Our work makes the following contributions: 1) We develop a policy that uses the perturbation in each value, weighted by the Fisher information, to select which weight and activation blocks to keep in higher precision. This approach preserves accuracy by identifying which weight and activation blocks need to be retained in higher precision to minimize the perturbation in the model loss. 2) We also propose a sensitivity-weighted clipping approach for fine-grained quantization which helps retain accuracy for blocks that are quantized to low precision. 3) We then propose hardware augmentations to leverage the efficiency benefits of FGMP quantization. Our hardware implementation encompasses i) datapath support for FGMP at block granularity, and ii) a mixed-precision activation quantization unit to assign activation blocks to high or low precision on the fly with minimal runtime and energy overhead. Our design, prototyped using NVFP4 (an FP4 format with microscaling) as the low-precision datatype and FP8 as the high-precision datatype, facilitates efficient FGMP quantization, attaining <1% perplexity degradation on Wikitext-103 for the Llama-2-7B model relative to an all-FP8 baseline design while consuming 14% less energy during inference and requiring 30% less weight memory.