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.

Zur Installation springen

Repository: davila7/claude-code-templates
Letzte Quellaktivität: 8. Januar 2026 um 14:24
Erkannte Sprache von SKILL.md: Englisch
Sterne: 30.229
Forks: 3.394

Installationsoptionen

Standardmäßig ist der Prompt ausgewählt, der zuerst die Quelle prüft. Sie können zu einem direkten Befehl wechseln oder eine lokale Kopie herunterladen.

Mit Codex oder Claude installieren Kopieren Sie diesen Prompt, fügen Sie ihn in Codex, Claude oder einen anderen Assistant ein und lassen Sie die Skill-Seite prüfen und installieren.

Ein direkter Befehl überspringt den Prüf-Prompt. Prüfen Sie die Quelle, bevor Sie ihn ausführen.

npx skills add https://github.com/davila7/claude-code-templates --skill gptq

Der Befehl bleibt in einer Zeile. Scrollen Sie horizontal, um ihn vor dem Kopieren vollständig zu prüfen.

Sie bevorzugen eine lokale Kopie? Laden Sie die Dateien herunter, die SkillsMP derzeit vorliegen.

Quelldateien prüfen

Lesen Sie SKILL.md und alle von SkillsMP angezeigten Begleitdateien, bevor Sie sich für eine Installation entscheiden.

Datei-Explorer

4 Dateien

SKILL.md

Quellanweisungen · Schreibgeschützte Vorschau

Zurück zur Suche Zurück zur Installation

computer-use-agents

Build AI agents that interact with computers like humans do - viewing screens, moving cursors, clicking buttons, and typing text. Covers Anthropic's Computer Use, OpenAI's Operator/CUA, and open-source alternatives. Critical focus on sandboxing, security, and handling the unique challenges of vision-based control. Use when: computer use, desktop automation agent, screen control AI, vision-based agent, GUI automation.

2026-08-04

ai-wrapper-product

Expert in building products that wrap AI APIs (OpenAI, Anthropic, etc.) into focused tools people will pay for. Not just 'ChatGPT but different' - products that solve specific problems with AI. Covers prompt engineering for products, cost management, rate limiting, and building defensible AI businesses. Use when: AI wrapper, GPT product, AI tool, wrap AI, AI SaaS.

2026-08-04

github-workflow-automation

Automate GitHub workflows with AI assistance. Includes PR reviews, issue triage, CI/CD integration, and Git operations. Use when automating GitHub workflows, setting up PR review automation, creating GitHub Actions, or triaging issues.

2026-08-04

Alle aus diesem Repository ansehen

# Install AutoGPTQ
pip install auto-gptq

# With Triton (Linux only, faster)
pip install auto-gptq[triton]

# With CUDA extensions (faster)
pip install auto-gptq --no-build-isolation

# Full installation
pip install auto-gptq transformers accelerate

from transformers import AutoTokenizer
from auto_gptq import AutoGPTQForCausalLM

# Load quantized model from HuggingFace
model_name = "TheBloke/Llama-2-7B-Chat-GPTQ"

model = AutoGPTQForCausalLM.from_quantized(
    model_name,
    device="cuda:0",
    use_triton=False  # Set True on Linux for speed
)

tokenizer = AutoTokenizer.from_pretrained(model_name)

# Generate
prompt = "Explain quantum computing"
inputs = tokenizer(prompt, return_tensors="pt").to("cuda:0")
outputs = model.generate(**inputs, max_new_tokens=200)
print(tokenizer.decode(outputs[0]))

from transformers import AutoTokenizer
from auto_gptq import AutoGPTQForCausalLM, BaseQuantizeConfig
from datasets import load_dataset

# Load model
model_name = "meta-llama/Llama-2-7b-chat-hf"
tokenizer = AutoTokenizer.from_pretrained(model_name)

# Quantization config
quantize_config = BaseQuantizeConfig(
    bits=4,              # 4-bit quantization
    group_size=128,      # Group size (recommended: 128)
    desc_act=False,      # Activation order (False for CUDA kernel)
    damp_percent=0.01    # Dampening factor
)

# Load model for quantization
model = AutoGPTQForCausalLM.from_pretrained(
    model_name,
    quantize_config=quantize_config
)

# Prepare calibration data
dataset = load_dataset("c4", split="train", streaming=True)
calibration_data = [
    tokenizer(example["text"])["input_ids"][:512]
    for example in dataset.take(128)
]

# Quantize
model.quantize(calibration_data)

# Save quantized model
model.save_quantized("llama-2-7b-gptq")
tokenizer.save_pretrained("llama-2-7b-gptq")

# Push to HuggingFace
model.push_to_hub("username/llama-2-7b-gptq")

Group Size	Model Size	Accuracy	Speed	Recommendation
-1 (per-column)	Smallest	Best	Slowest	Research only
32	Smaller	Better	Slower	High accuracy needed
128	Medium	Good	Fast	Recommended default
256	Larger	Lower	Faster	Speed critical
1024	Largest	Lowest	Fastest	Not recommended

Weight matrix: [1024, 4096] = 4.2M elements

Group size = 128:
- Groups: 4.2M / 128 = 32,768 groups
- Each group: own 4-bit scale + zero-point
- Result: Better granularity → better accuracy

from auto_gptq import BaseQuantizeConfig

config = BaseQuantizeConfig(
    bits=4,              # 4-bit quantization
    group_size=128,      # Standard group size
    desc_act=False,      # Faster CUDA kernel
    damp_percent=0.01    # Dampening factor
)

config = BaseQuantizeConfig(
    bits=3,              # 3-bit (more compression)
    group_size=128,      # Keep standard group size
    desc_act=True,       # Better accuracy (slower)
    damp_percent=0.01
)

config = BaseQuantizeConfig(
    bits=4,
    group_size=32,       # Smaller groups (better accuracy)
    desc_act=True,       # Activation reordering
    damp_percent=0.005   # Lower dampening
)

model = AutoGPTQForCausalLM.from_quantized(
    model_name,
    device="cuda:0",
    use_exllama=True,      # Use ExLlamaV2
    exllama_config={"version": 2}
)

# Quantize with Marlin format
config = BaseQuantizeConfig(
    bits=4,
    group_size=128,
    desc_act=False  # Required for Marlin
)

model.quantize(calibration_data, use_marlin=True)

# Load with Marlin
model = AutoGPTQForCausalLM.from_quantized(
    model_name,
    device="cuda:0",
    use_marlin=True  # 2× faster on A100/H100
)

model = AutoGPTQForCausalLM.from_quantized(
    model_name,
    device="cuda:0",
    use_triton=True  # Linux only
)

from transformers import AutoModelForCausalLM, AutoTokenizer

# Load quantized model (transformers auto-detects GPTQ)
model = AutoModelForCausalLM.from_pretrained(
    "TheBloke/Llama-2-13B-Chat-GPTQ",
    device_map="auto",
    trust_remote_code=False
)

tokenizer = AutoTokenizer.from_pretrained("TheBloke/Llama-2-13B-Chat-GPTQ")

# Use like any transformers model
inputs = tokenizer("Hello", return_tensors="pt").to("cuda")
outputs = model.generate(**inputs, max_new_tokens=100)

from transformers import AutoModelForCausalLM
from peft import prepare_model_for_kbit_training, LoraConfig, get_peft_model

# Load GPTQ model
model = AutoModelForCausalLM.from_pretrained(
    "TheBloke/Llama-2-7B-GPTQ",
    device_map="auto"
)

# Prepare for LoRA training
model = prepare_model_for_kbit_training(model)

# LoRA config
lora_config = LoraConfig(
    r=16,
    lora_alpha=32,
    target_modules=["q_proj", "v_proj"],
    lora_dropout=0.05,
    bias="none",
    task_type="CAUSAL_LM"
)

# Add LoRA adapters
model = get_peft_model(model, lora_config)

# Fine-tune (memory efficient!)
# 70B model trainable on single A100 80GB

# Automatic device mapping
model = AutoGPTQForCausalLM.from_quantized(
    "TheBloke/Llama-2-70B-GPTQ",
    device_map="auto",  # Automatically split across GPUs
    max_memory={0: "40GB", 1: "40GB"}  # Limit per GPU
)

# Manual device mapping
device_map = {
    "model.embed_tokens": 0,
    "model.layers.0-39": 0,  # First 40 layers on GPU 0
    "model.layers.40-79": 1,  # Last 40 layers on GPU 1
    "model.norm": 1,
    "lm_head": 1
}

model = AutoGPTQForCausalLM.from_quantized(
    model_name,
    device_map=device_map
)

# Offload some layers to CPU (for very large models)
model = AutoGPTQForCausalLM.from_quantized(
    "TheBloke/Llama-2-405B-GPTQ",
    device_map="auto",
    max_memory={
        0: "80GB",  # GPU 0
        1: "80GB",  # GPU 1
        2: "80GB",  # GPU 2
        "cpu": "200GB"  # Offload overflow to CPU
    }
)

# Process multiple prompts efficiently
prompts = [
    "Explain AI",
    "Explain ML",
    "Explain DL"
]

inputs = tokenizer(prompts, return_tensors="pt", padding=True).to("cuda")

outputs = model.generate(
    **inputs,
    max_new_tokens=100,
    pad_token_id=tokenizer.eos_token_id
)

for i, output in enumerate(outputs):
    print(f"Prompt {i}: {tokenizer.decode(output)}")

# Find GPTQ models on HuggingFace
https://huggingface.co/models?library=gptq

from auto_gptq import AutoGPTQForCausalLM

# Automatically downloads from HuggingFace
model = AutoGPTQForCausalLM.from_quantized(
    "TheBloke/Llama-2-70B-Chat-GPTQ",
    device="cuda:0"
)

name	gptq
description	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.
version	1.0.0
author	Orchestra Research
license	MIT
tags	["Optimization","GPTQ","Quantization","4-Bit","Post-Training","Memory Optimization","Consumer GPUs","Fast Inference","QLoRA","Group-Wise Quantization"]
dependencies	["auto-gptq","transformers","optimum","peft"]

name	gptq
description	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.
version	1.0.0
author	Orchestra Research
license	MIT
tags	["Optimization","GPTQ","Quantization","4-Bit","Post-Training","Memory Optimization","Consumer GPUs","Fast Inference","QLoRA","Group-Wise Quantization"]
dependencies	["auto-gptq","transformers","optimum","peft"]

Model	FP16	GPTQ 4-bit	Reduction
Llama 2-7B	14 GB	3.5 GB	4×
Llama 2-13B	26 GB	6.5 GB	4×
Llama 2-70B	140 GB	35 GB	4×
Llama 3-405B	810 GB	203 GB	4×

Precision	Tokens/sec	vs FP16
FP16	25 tok/s	1×
GPTQ 4-bit (CUDA)	85 tok/s	3.4×
GPTQ 4-bit (ExLlama)	105 tok/s	4.2×
GPTQ 4-bit (Marlin)	120 tok/s	4.8×

Model	FP16	GPTQ 4-bit (g=128)	Degradation
Llama 2-7B	5.47	5.55	+1.5%
Llama 2-13B	4.88	4.95	+1.4%
Llama 2-70B	3.32	3.38	+1.8%

gptq

Installationsoptionen

Quelldateien prüfen

GPTQ (Generative Pre-trained Transformer Quantization)

When to use GPTQ

GPTQ (Generative Pre-trained Transformer Quantization)

When to use GPTQ

Quick start

Installation

Load pre-quantized model

Quantize your own model

Group-wise quantization

Quantization configurations

Standard 4-bit (recommended)

High accuracy (3-bit with larger groups)

Maximum accuracy (4-bit with small groups)

Kernel backends

ExLlamaV2 (default, fastest)

Marlin (Ampere+ GPUs)

Triton (Linux only)

Integration with transformers

Direct transformers usage

QLoRA fine-tuning (GPTQ + LoRA)

Performance benchmarks

Memory reduction

Inference speed (Llama 2-7B, A100)

Accuracy (perplexity on WikiText-2)

Common patterns

Multi-GPU deployment

CPU offloading

Batch inference

Finding pre-quantized models

Supported models

References

Resources

gptq

Quellinformationen

Installationsoptionen

Quelldateien prüfen

Mehr aus diesem Repository

Mehr aus diesem Repository

GPTQ (Generative Pre-trained Transformer Quantization)

When to use GPTQ

GPTQ (Generative Pre-trained Transformer Quantization)

When to use GPTQ

Quick start

Installation

Load pre-quantized model

Quantize your own model

Group-wise quantization

Quantization configurations

Standard 4-bit (recommended)

High accuracy (3-bit with larger groups)

Maximum accuracy (4-bit with small groups)

Kernel backends

ExLlamaV2 (default, fastest)

Marlin (Ampere+ GPUs)

Triton (Linux only)

Integration with transformers

Direct transformers usage

QLoRA fine-tuning (GPTQ + LoRA)

Performance benchmarks

Memory reduction

Inference speed (Llama 2-7B, A100)

Accuracy (perplexity on WikiText-2)

Common patterns

Multi-GPU deployment

CPU offloading

Batch inference

Finding pre-quantized models

Supported models

References

Resources