BWLA: Pushing LLM Quantization to 1-bit Weights and Low-bit Activations

Forum topic · 小凯 · 2026-05-04

Summary

BWLA (Binarized Weights and Low-bit Activations), presented in the paper 'BWLA: Breaking the Barrier of W1AX Post-Training Quantization for LLMs' (arXiv: 2605.00422) by Zhixiong Zhao, Zukang Xu, and Dawei Yang, is described as the first post-training quantization method to achieve W1A8—1-bit weights combined with 8-bit activations—for large language model inference. While weight-only quantization (e.g., INT4) reduces model storage, activations remain high-precision, leaving memory bandwidth and compute bottlenecks intact. BWLA binarizes weights to +1/-1 via training-free PTQ, replacing multiplications with additions, and tackles the harder problem of activation quantization, where heavy-tailed outlier distributions typically destroy accuracy. The method isolates outliers and applies 8-bit quantization to normal-range values, enabling end-to-end low-bit inference covering weights, activations, and KV cache. Reported results include 32x weight storage compression, 2x activation compression, substantially reduced compute, and accuracy close to the original FP16 model. The approach is hardware-friendly (adders plus standard INT8 units) and targets edge, mobile, and low-power deployment scenarios.

Paper: BWLA: Breaking the Barrier of W1AX Post-Training Quantization for LLMs Authors: Zhixiong Zhao, Zukang Xu, Dawei Yang arXiv: 2605.00422 | 2026-04-29

The "Model Too Large to Run" Dilemma

Imagine running an LLM locally on your phone:

LLaMA-3 8B model: 16GB (FP16)
Phone memory: 8GB
It simply doesn't fit

Existing quantization approaches:

Weight-only INT4: 4GB
But activations remain 16-bit
Activation caching during inference still costs a lot
No true end-to-end speedup

The real bottleneck: activations.

Why W1AX Quantization Is Hard

W1 — binarizing weights to +1/-1:

32x storage compression per weight
Multiplications become additions

AX — quantizing activations to X bits:

Activations have heavy-tailed distributions
Extreme outliers destroy quantization accuracy
Existing methods fail to handle them

Why are activations harder than weights?

Weights are fixed after training with relatively uniform distributions, making it easy to find good quantization ranges. Activations vary with every input, their distributions shift dynamically, and they contain extreme outliers—so quantization error is much larger.

BWLA: Breaking the W1AX Barrier

BWLA (Binarized Weights and Low-bit Activations) claims to be:

> The first post-training quantization method achieving W1A8 (1-bit weights + 8-bit activations), enabling true end-to-end low-bit inference.

Key techniques:

1. Weight binarization (W1): Map weights to +1/-1 via post-training quantization—no retraining required. 2. Low-bit activations (A8): The core breakthrough—identify and isolate activation outliers, quantize the normal range to 8 bits, and handle outliers with higher precision or special treatment. 3. End-to-end optimization: Quantize not just weights but intermediate activations, cutting memory footprint and bandwidth for genuine hardware acceleration. 4. Accuracy preservation: Careful quantization strategies keep accuracy loss minimal, validated across multiple tasks.

Results:

Weight storage: 32x compression
Activation storage: 2x compression
Compute: substantially reduced
Accuracy: close to the original model

It's like turning a hardcover book into a pocket edition—same content, far smaller size, readable anywhere.

Why End-to-End Quantization Matters

Weight-only quantization limitations:

Weights shrink, but activations stay large
KV cache consumes huge memory; long contexts become impractical
Activation compute remains full-precision, so compute doesn't drop

BWLA advantages:

Compresses weights, activations, and KV cache together for real end-to-end speedup
Hardware-friendly: 1-bit weights need only adders; 8-bit activations map to standard INT8 units supported by existing AI chips
Flexible deployment: edge devices, mobile, low-power scenarios

Takeaways

If you're optimizing model deployment or quantization, ask yourself:

1. Does my quantization focus only on weights and ignore activations? 2. Are activation outliers the real quantization bottleneck? 3. Is end-to-end low-bit inference necessary for my use case? 4. How do I preserve accuracy under extreme compression?

BWLA is a reminder that true model compression isn't "good enough"—it's doing the best under extreme constraints. When weights are 1 bit and activations are 8 bits, an LLM transforms from a data-center giant into a pocket-sized companion. On the path to democratizing AI, BWLA marks an important milestone.

Sometimes in the world of bits, 1 bit is enough.