
BitsAndBytes Quantization
FreeOptimize LLMs with efficient quantization techniques.
Free · Opens the source repo
What BitsAndBytes Quantization does
BitsAndBytes is a powerful tool designed for quantizing large language models (LLMs) to 8-bit or 4-bit formats, achieving significant memory savings while maintaining accuracy. This skill is particularly useful for developers and data scientists working with large models who face constraints on GPU memory. By reducing memory usage by 50% with 8-bit quantization and up to 75% with 4-bit quantization, BitsAndBytes allows users to load larger models or perform faster inference without a substantial drop in performance.
The skill integrates seamlessly with HuggingFace Transformers, making it easy to implement quantization in existing workflows. Users can quickly set up their models with minimal configuration, utilizing the provided code snippets to load models in either 8-bit or 4-bit formats. The tool also supports various quantization formats, including INT8 and NF4, enabling users to tailor their approach based on specific model requirements and hardware capabilities.
In addition to quantization, BitsAndBytes includes support for QLoRA training, which allows for fine-tuning large models on consumer-grade GPUs. This feature is essential for those looking to adapt pre-trained models to specific tasks without needing extensive computational resources. The skill also offers an 8-bit optimizer, which can further reduce memory usage during training, making it a comprehensive solution for memory optimization in deep learning workflows.
Overall, BitsAndBytes is an essential tool for anyone working with large language models who needs to optimize memory usage while preserving model performance. It simplifies the process of quantization and fine-tuning, making advanced machine learning techniques accessible even on limited hardware.
When to use it
Use BitsAndBytes when you need to load large models in environments with constrained GPU memory or when you want to accelerate inference times.
When not to use it
This skill may not be suitable for scenarios where maximum model accuracy is critical, as quantization can introduce minor accuracy loss, or if you are not working with large language models that require such optimizations.
What you can build with it
Load Large Models on Limited Hardware
When working with a GPU that has limited memory, BitsAndBytes allows you to load larger models by quantizing them to 8-bit or 4-bit formats.
Fine-tune Models with QLoRA
Use BitsAndBytes to fine-tune large language models efficiently on consumer-grade GPUs, utilizing the QLoRA framework.
Optimize Training with 8-bit Optimizers
Incorporate 8-bit optimizers into your training routine to significantly reduce memory usage while maintaining training performance.
How to install BitsAndBytes Quantization
View source1. Install with the skills CLI
npx skills add davila7/claude-code-templates/optimization-bitsandbytes --agent claude-code2. Or install it manually
Download the skill folder and drop it into ~/.claude/skills/ for all projects, or .claude/skills/ to scope it to one repo. Restart Claude Code so it picks up the new skill.
Anthropic's agentic coding CLI, and the reference implementation of Agent Skills. Drop a skill folder into ~/.claude/skills and Claude Code loads it automatically whenever a task matches the skill's description. Claude Code docs
Inside SKILL.md
Written by davila7bitsandbytes - LLM Quantization
Quick start
bitsandbytes reduces LLM memory by 50% (8-bit) or 75% (4-bit) with <1% accuracy loss.
Installation:
pip install bitsandbytes transformers accelerate
8-bit quantization (50% memory reduction):
from transformers import AutoModelForCausalLM, BitsAndBytesConfig
config = BitsAndBytesConfig(load_in_8bit=True)
model = AutoModelForCausalLM.from_pretrained(
"meta-llama/Llama-2-7b-hf",
quantization_config=config,
device_map="auto"
)
# Memory: 14GB → 7GB
4-bit quantization (75% memory reduction):
config = BitsAndBytesConfig(
load_in_4bit=True,
bnb_4bit_compute_dtype=torch.float16
)
model = AutoModelForCausalLM.from_pretrained(
"meta-llama/Llama-2-7b-hf",
quantization_config=config,
device_map="auto"
)
# Memory: 14GB → 3.5GB
Common workflows
Workflow 1: Load large model in limited GPU memory
Copy this checklist:
Quantization Loading:
- [ ] Step 1: Calculate memory requirements
- [ ] Step 2: Choose quantization level (4-bit or 8-bit)
- [ ] Step 3: Configure quantization
- [ ] Step 4: Load and verify model
Step 1: Calculate memory requirements
Estimate model memory:
FP16 memory (GB) = Parameters × 2 bytes / 1e9
INT8 memory (GB) = Parameters × 1 byte / 1e9
INT4 memory (GB) = Parameters × 0.5 bytes / 1e9
Example (Llama 2 7B):
FP16: 7B × 2 / 1e9 = 14 GB
INT8: 7B × 1 / 1e9 = 7 GB
INT4: 7B × 0.5 / 1e9 = 3.5 GB
Step 2: Choose quantization level
| GPU VRAM | Model Size | Recommended |
|---|---|---|
| 8 GB | 3B | 4-bit |
| 12 GB | 7B | 4-bit |
| 16 GB | 7B | 8-bit or 4-bit |
| 24 GB | 13B | 8-bit or 70B 4-bit |
| 40+ GB | 70B | 8-bit |
Step 3: Configure quantization
For 8-bit (better accuracy):
from transformers import BitsAndBytesConfig
import torch
config = BitsAndBytesConfig(
load_in_8bit=True,
llm_int8_threshold=6.0, # Outlier threshold
llm_int8_has_fp16_weight=False
)
For 4-bit (maximum memory savings):
config = BitsAndBytesConfig(
load_in_4bit=True,
bnb_4bit_compute_dtype=torch.float16, # Compute in FP16
bnb_4bit_quant_type="nf4", # NormalFloat4 (recommended)
bnb_4bit_use_double_quant=True # Nested quantization
)
Step 4: Load and verify model
from transformers import AutoModelForCausalLM, AutoTokenizer
model = AutoModelForCausalLM.from_pretrained(
"meta-llama/Llama-2-13b-hf",
quantization_config=config,
device_map="auto", # Automatic device placement
torch_dtype=torch.float16
)
tokenizer = AutoTokenizer.from_pretrained("meta-llama/Llama-2-13b-hf")
# Test inference
inputs = tokenizer("Hello, how are you?", return_tensors="pt").to("cuda")
outputs = model.generate(**inputs, max_length=50)
print(tokenizer.decode(outputs[0]))
# Check memory
import torch
print(f"Memory allocated: {torch.cuda.memory_allocated()/1e9:.2f}GB")
Workflow 2: Fine-tune with QLoRA (4-bit training)
QLoRA enables fine-tuning large models on consumer GPUs.
Copy this checklist:
QLoRA Fine-tuning:
- [ ] Step 1: Install dependencies
- [ ] Step 2: Configure 4-bit base model
- [ ] Step 3: Add LoRA adapters
- [ ] Step 4: Train with standard Trainer
Step 1: Install dependencies
pip install bitsandbytes transformers peft accelerate datasets
Step 2: Configure 4-bit base model
from transformers import AutoModelForCausalLM, BitsAndBytesConfig
import torch
bnb_config = BitsAndBytesConfig(
load_in_4bit=True,
bnb_4bit_compute_dtype=torch.float16,
bnb_4bit_quant_type="nf4",
bnb_4bit_use_double_quant=True
)
model = AutoModelForCausalLM.from_pretrained(
"meta-llama/Llama-2-7b-hf",
quantization_config=bnb_config,
device_map="auto"
)
Step 3: Add LoRA adapters
from peft import LoraConfig, get_peft_model, prepare_model_for_kbit_training
# Prepare model for training
model = prepare_model_for_kbit_training(model)
# Configure LoRA
lora_config = LoraConfig(
r=16, # LoRA rank
lora_alpha=32, # LoRA alpha
target_modules=["q_proj", "k_proj", "v_proj", "o_proj"],
lora_dropout=0.05,
bias="none",
task_type="CAUSAL_LM"
)
# Add LoRA adapters
model = get_peft_model(model, lora_config)
model.print_trainable_parameters()
# Output: trainable params: 4.2M || all params: 6.7B || trainable%: 0.06%
Step 4: Train with standard Trainer
from transformers import Trainer, TrainingArguments
training_args = TrainingArguments(
output_dir="./qlora-output",
per_device_train_batch_size=4,
gradient_accumulation_steps=4,
num_train_epochs=3,
learning_rate=2e-4,
fp16=True,
logging_steps=10,
save_strategy="epoch"
)
trainer = Trainer(
model=model,
args=training_args,
train_dataset=train_dataset,
tokenizer=tokenizer
)
trainer.train()
# Save LoRA adapters (only ~20MB)
model.save_pretrained("./qlora-adapters")
Workflow 3: 8-bit optimizer for memory-efficient training
Use 8-bit Adam/AdamW to reduce optimizer memory by 75%.
8-bit Optimizer Setup:
- [ ] Step 1: Replace standard optimizer
- [ ] Step 2: Configure training
- [ ] Step 3: Monitor memory savings
Step 1: Replace standard optimizer
import bitsandbytes as bnb
from transformers import Trainer, TrainingArguments
# Instead of torch.optim.AdamW
model = AutoModelForCausalLM.from_pretrained("model-name")
training_args = TrainingArguments(
output_dir="./output",
per_device_train_batch_size=8,
optim="paged_adamw_8bit", # 8-bit optimizer
learning_rate=5e-5
)
trainer = Trainer(
model=model,
args=training_args,
train_dataset=train_dataset
)
trainer.train()
Manual optimizer usage:
import bitsandbytes as bnb
optimizer = bnb.optim.AdamW8bit(
model.parameters(),
lr=1e-4,
betas=(0.9, 0.999),
eps=1e-8
)
# Training loop
for batch in dataloader:
loss = model(**batch).loss
loss.backward()
optimizer.step()
optimizer.zero_grad()
Step 2: Configure training
Compare memory:
Standard AdamW optimizer memory = model_params × 8 bytes (states)
8-bit AdamW memory = model_params × 2 bytes
Savings = 75% optimizer memory
Example (Llama 2 7B):
Standard: 7B × 8 = 56 GB
8-bit: 7B × 2 = 14 GB
Savings: 42 GB
Step 3: Monitor memory savings
import torch
before = torch.cuda.memory_allocated()
# Training step
optimizer.step()
after = torch.cuda.memory_allocated()
print(f"Memory used: {(after-before)/1e9:.2f}GB")
When to use vs alternatives
Use bitsandbytes when:
- GPU memory limited (need to fit larger model)
- Training with QLoRA (fine-tune 70B on single GPU)
- Inference only (50-75% memory reduction)
- Using HuggingFace Transformers
- Acceptable 0-2% accuracy degradation
Use alternatives instead:
- GPTQ/AWQ: Production serving (faster inference than bitsandbytes)
- GGUF: CPU inference (llama.cpp)
- FP8: H100 GPUs (hardware FP8 faster)
- Full precision: Accuracy critical, memory not constrained
Common issues
Issue: CUDA error during loading
Install matching CUDA version:
# Check CUDA version
nvcc --version
# Install matching bitsandbytes
pip install bitsandbytes --no-cache-dir
Issue: Model loading slow
Use CPU offload for large models:
model = AutoModelForCausalLM.from_pretrained(
"model-name",
quantization_config=config,
device_map="auto",
max_memory={0: "20GB", "cpu": "30GB"} # Offload to CPU
)
Issue: Lower accuracy than expected
Try 8-bit instead of 4-bit:
config = BitsAndBytesConfig(load_in_8bit=True)
# 8-bit has <0.5% accuracy loss vs 1-2% for 4-bit
Or use NF4 with double quantization:
config = BitsAndBytesConfig(
load_in_4bit=True,
bnb_4bit_quant_type="nf4", # Better than fp4
bnb_4bit_use_double_quant=True # Extra accuracy
)
Issue: OOM even with 4-bit
Enable CPU offload:
model = AutoModelForCausalLM.from_pretrained(
"model-name",
quantization_config=config,
device_map="auto",
offload_folder="offload", # Disk offload
offload_state_dict=True
)
Advanced topics
QLoRA training guide: See references/qlora-training.md for complete fine-tuning workflows, hyperparameter tuning, and multi-GPU training.
Quantization formats: See references/quantization-formats.md for INT8, NF4, FP4 comparison, double quantization, and custom quantization configs.
Memory optimization: See references/memory-optimization.md for CPU offloading strategies, gradient checkpointing, and memory profiling.
Hardware requirements
- GPU: NVIDIA with compute capability 7.0+ (Turing, Ampere, Hopper)
- VRAM: Depends on model and quantization
- 4-bit Llama 2 7B: 4GB
- 4-bit Llama 2 13B: 8GB
- 4-bit Llama 2 70B: 24GB
- CUDA: 11.1+ (12.0+ recommended)
- PyTorch: 2.0+
Supported platforms: NVIDIA GPUs (primary), AMD ROCm, Intel GPUs (experimental)
Resources
- GitHub: https://github.com/bitsandbytes-foundation/bitsandbytes
- HuggingFace docs: https://huggingface.co/docs/transformers/quantization/bitsandbytes
- QLoRA paper: "QLoRA: Efficient Finetuning of Quantized LLMs" (2023)
- LLM.int8() paper: "LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scale" (2022)
Frequently asked questions about BitsAndBytes Quantization
Similar skills
Heap Snapshot Analysis
Investigate V8 heap snapshots for memory issues.
VS Code Performance Workflow
Automate performance investigations in VS Code.
Memory Leak Audit
Prevent memory leaks with effective coding patterns.
CPU Profile Analysis
Analyze V8 and Chrome performance profiles for optimization.
Chat Performance Testing
Benchmark and validate chat UI performance in VS Code.
Vercel React Best Practices
Optimize your React and Next.js applications for performance.
