research-hqq

A guide to Half-Quadratic Quantization (HQQ), a method for reducing the number of bits used to store large language model weights without calibration data.

In plain words
What is it for?
Use it to quantize models to 8-, 4-, 3-, 2-, or 1-bit precision, deploy them with supported PyTorch, Hugging Face, or vLLM workflows, or fine-tune quantized models with LoRA or PEFT.
Why use it?
It helps make models smaller and supports fast quantization workflows when you do not have a calibration dataset.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/graycodeai/starling/research-hqq
Any agent
npx skills add GrayCodeAI/starling --skill research-hqq
Clone the repo
git clone --depth 1 https://github.com/GrayCodeAI/starling

Made for: Claude Code, Codex.

Per session 48 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,194 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin 97% copy Near-identical to another mod in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00048 $0.03194
Opus 5 $0.00024 $0.01597
Sonnet 5 $0.00010 $0.00639
Haiku 4.5 $0.00005 $0.00319

Measured 2d ago against content hash d4798d739bc2, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

research-hqq scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

Origin

This is a copy

97% identical to hqq-quantization — 13 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

categories/ai-ml/research-hqq/SKILL.md · 445 lines

How it starts

The opening of the file, as written. The whole thing — 445 lines — stays where its author put it; the contents beside it link to each section on GitHub.

HQQ - Half-Quadratic Quantization

Fast, calibration-free weight quantization supporting 8/4/3/2/1-bit precision with multiple optimized backends.

When to use HQQ

Use HQQ when:

  • Quantizing models without calibration data (no dataset needed)
  • Need fast quantization (minutes vs hours for GPTQ/AWQ)
  • Deploying with vLLM or HuggingFace Transformers
  • Fine-tuning quantized models with LoRA/PEFT
  • Experimenting with extreme quantization (2-bit, 1-bit)

Key advantages:

  • No calibration: Quantize any model instantly without sample data
  • Multiple backends: PyTorch, ATEN, TorchAO, Marlin, BitBlas for optimized inference
  • Flexible precision: 8/4/3/2/1-bit with configurable group sizes
  • Framework integration: Native HuggingFace and vLLM support
  • PEFT compatible: Fine-tune quantized models with LoRA

Use alternatives instead:

  • AWQ: Need calibration-based accuracy, production serving
  • GPTQ: Maximum accuracy with calibration data available
  • bitsandbytes: Simple 8-bit/4-bit without custom backends
  • llama.cpp/GGUF: CPU inference, Apple Silicon deployment

Quick start

Installation

pip install hqq

# With specific backend
pip install hqq[torch]      # PyTorch backend
pip install hqq[torchao]    # TorchAO int4 backend
pip install hqq[bitblas]    # BitBlas backend
pip install hqq[marlin]     # Marlin backend

Basic quantization

from hqq.core.quantize import BaseQuantizeConfig, HQQLinear
import torch.nn as nn

# Configure quantization
config = BaseQuantizeConfig(
    nbits=4,           # 4-bit quantization
    group_size=64,     # Group size for quantization
    axis=1             # Quantize along output dimension
)

# Quantize a linear layer
linear = nn.Linear(4096, 4096)
hqq_linear = HQQLinear(linear, config)

# Use normally
output = hqq_linear(input_tensor)

Quantize full model with HuggingFace

from transformers import AutoModelForCausalLM, HqqConfig

# Configure HQQ
quantization_config = HqqConfig(
    nbits=4,
    group_size=64,
    axis=1
)

# Load and quantize
model = AutoModelForCausalLM.from_pretrained(
    "meta-llama/Llama-3.1-8B",
    quantization_config=quantization_config,
    device_map="auto"
)

# Model is quantized and ready to use

Read the full file on GitHub · 445 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 445 lines · 48 tokens per session scan A d4798d739bc2

Subscribe to this mod's changes

research-hqq is a skill published in the GitHub repository GrayCodeAI/starling (2 stars, last pushed 2d ago), licensed MIT. It adds 48 tokens to every session and 3,194 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. It is 97% identical to hqq-quantization, differing in 13 lines, and is treated as a copy.

Related

Other skills, from other repositories

hqq-quantization

Half-Quadratic Quantization for LLMs without calibration data. Use when quantizing models to 4/3/2-bit precision without needing calibration datasets, for fast quantization workflows, or when deploying with vLLM or HuggingFace Transformers.

Orchestra-Research/AI-Research-SKILLs · 58 tokens

hqq-quantization

Half-Quadratic Quantization for LLMs without calibration data. Use when quantizing models to 4/3/2-bit precision without needing calibration datasets, for fast quantization workflows, or when deploying with vLLM or HuggingFace Transformers.

ihatesea69/HieuNghi-AI-Skills · 58 tokens

gptq

Post-training 4-bit quantization for LLMs with minimal accuracy loss. Use for deploying large models (70B, 405B) on consumer GPUs, when you need 4× memory reduction with <2% perplexity degradation, or for faster inference (3-4× speedup) vs FP16. Integrates with transformers and PEFT for QLoRA fine-tuning.

Orchestra-Research/AI-Research-SKILLs · 84 tokens

awq-quantization

Activation-aware weight quantization for 4-bit LLM compression with 3x speedup and minimal accuracy loss. Use when deploying large models (7B-70B) on limited GPU memory, when you need faster inference than GPTQ with better accuracy preservation, or for instruction-tuned and multimodal models. MLSys 2024 Best Paper…

Orchestra-Research/AI-Research-SKILLs · 79 tokens

quantizing-models-bitsandbytes

Quantizes LLMs to 8-bit or 4-bit for 50-75% memory reduction with minimal accuracy loss. Use when GPU memory is limited, need to fit larger models, or want faster inference. Supports INT8, NF4, FP4 formats, QLoRA training, and 8-bit optimizers. Works with HuggingFace Transformers.

Orchestra-Research/AI-Research-SKILLs · 83 tokens

optimizing-attention-flash

Optimizes transformer attention with Flash Attention for 2-4x speedup and 10-20x memory reduction. Use when training/running transformers with long sequences (>512 tokens), encountering GPU memory issues with attention, or need faster inference. Supports PyTorch native SDPA, flash-attn library, H100 FP8, and sliding…

Orchestra-Research/AI-Research-SKILLs · 78 tokens