Mapping the Schedule × Bit-Width Boundary in Sub-100M Quantisation-Aware Training
A large factorial grid study (1345 total runs across two phases) tests whether optimal learning-rate schedules differ by bit-width during from-scratch quantisation-aware training (QAT) for sub-100M decoder language models. The primary hypothesis—that INT6 QAT requires a different schedule than FP16/INT8—is falsified; a 33% warmdown fraction is optimal across all precisions and model sizes from 5M to 350M. For INT4, a regime boundary is identified near 50M parameters: above it, wd33 is decisively optimal; below it, schedule choice falls within seed-level noise. The study also establishes a log-linear scaling law for the INT6 quantisation penalty that successfully predicts held-out model sizes.
Related guides (3)
Related events (8)
Fine-tuning LLMs to 1.58bit: extreme quantization made easy
Hugging Face published a blog post describing a method for fine-tuning large language models down to 1.58-bit precision, referencing the BitNet b1.58 quantization scheme. The post covers tooling and workflows that make extreme quantization more accessible via the Hugging Face ecosystem. This represents a practical guide to applying ternary-weight quantization ({-1, 0, 1}) to existing models through fine-tuning rather than training from scratch.
Unlocking Longer Generation with Key-Value Cache Quantization
This Hugging Face blog post covers KV cache quantization as a technique to reduce memory consumption during LLM inference, enabling longer context generation without proportional VRAM increases. The post likely explains how quantizing the key-value cache (e.g., to INT8 or lower precision) trades minimal accuracy for significant memory savings. This is directly relevant to inference efficiency and long-context deployment patterns.
Ω-QVLA: Training-Free W4A4 Quantization for Full Vision-Language-Action Models Including Diffusion Action Heads
Omega-QVLA is a post-training quantization framework that compresses both the LLM backbone and the diffusion-based action head of VLA models to uniform W4A4 precision without mixed-precision schemes or fine-tuning. It combines composite SVD-Hadamard rotation for weight energy equalization with per-step DiT activation scaling to handle dynamic-range drift across denoising steps. On the LIBERO benchmark, it achieves 98.0% and 87.8% task success on Pi 0.5 and GR00T N1.5 respectively—matching or exceeding FP16 baselines—while reducing static memory footprint by 71.3%. Real-world manipulation experiments confirm the approach generalizes beyond simulation.
Variable-Width Transformers: X-shaped architecture outperforms uniform-width baselines with 22% fewer FLOPs
Researchers propose the ><former (X-shaped transformer), a decoder-only architecture that uses wider early and late layers with narrower middle layers, implemented via a parameter-free residual resizing mechanism. Evaluated on models from 200M to 2B dense parameters and 3B MoE, the architecture consistently outperforms parameter-matched uniform-width baselines on language modeling loss. The design yields a 22% reduction in FLOPs and 15% reduction in KV cache memory under fitted scaling curves, suggesting nonuniform width allocation is a viable path to more compute-efficient language models.
Making LLMs even more accessible with bitsandbytes, 4-bit quantization and QLoRA
Hugging Face published a blog post detailing the integration of 4-bit quantization via bitsandbytes into the Transformers library, enabling large language models to run on consumer-grade hardware. The post covers NF4 (NormalFloat4) data type and double quantization techniques from the QLoRA paper, which together reduce memory footprint significantly while preserving model quality. It demonstrates how users can load models like LLaMA in 4-bit precision and fine-tune them using QLoRA with minimal code changes.
Overview of Natively Supported Quantization Schemes in 🤗 Transformers
This Hugging Face blog post surveys the quantization methods natively integrated into the Transformers library as of September 2023, covering schemes such as GPTQ, bitsandbytes (LLM.int8, NF4), and related techniques. It explains how each method works, their trade-offs in terms of memory reduction and inference speed, and how practitioners can apply them via the Transformers API. The post serves as a practical reference for deploying large language models under memory constraints.
Q8-Chat: Efficient Generative AI on Intel Xeon via INT8 Quantization
Hugging Face and Intel demonstrate running quantized large language models (INT8/Q8) on Intel Xeon CPUs, branded as Q8-Chat. The post covers inference performance of quantized models on CPU hardware without requiring GPUs. This is relevant to inference economics and enterprise deployment, particularly for organizations without GPU infrastructure.
Introducing AutoRound: Intel's Advanced Quantization for LLMs and VLMs
Intel has released AutoRound, an advanced quantization technique for large language models and vision-language models, announced via the Hugging Face blog. AutoRound targets efficient low-bit quantization to reduce model size and inference costs while preserving accuracy. The tool is positioned as a production-ready quantization solution integrated with the Hugging Face ecosystem.