wpferrell/qwen2.5-14b-instruct-bigsmall

Qwen 2.5 14B Instruct (BigSmall compressed)

29.5 GB → 19.5 GB (66.1%). Under 2 GB peak RAM. Full quality — not quantization.

This is Qwen2.5-14B-Instruct compressed with BigSmall — lossless neural network weight compression. Every weight is bit-identical to the original. No accuracy loss whatsoever.

Install

pip install bigsmall

Load and run inference (streaming — under 2GB peak RAM)

from bigsmall import StreamingLoader
from transformers import AutoModelForCausalLM, AutoTokenizer

loader = StreamingLoader("wpferrell/qwen2.5-14b-instruct-bigsmall")
model = loader.load_model(AutoModelForCausalLM)
tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen2.5-14B-Instruct")

messages = [{"role": "user", "content": "Explain lossless compression."}]
text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tokenizer([text], return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=200)
print(tokenizer.decode(outputs[0]))

Or use AutoModel with the transparent hook

import bigsmall
bigsmall.install_hook()

from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("wpferrell/qwen2.5-14b-instruct-bigsmall")

Compression stats

Metric	Value
Original size	29.5 GB
Compressed size	19.5 GB
Ratio	66.1% (BF16)
Format	BF16 → BigSmall (.bs shards)
Lossless verified	md5 every tensor
Peak RAM (streaming)	< 2 GB

Comparison

Tool	BF16 Ratio	FP32 Ratio	Inference Overhead	Hardware
ZipNN	67%	83%	None	CPU
DFloat11	~70%	BF16 only	~2x at batch=1	CUDA only
ZipServ	~70%	BF16 only	1.22x faster	GDDR GPU
BigSmall	65.6%	75.5%	None	CPU + any GPU

About BigSmall

BigSmall compresses at the joint entropy floor for neural network weights. It codes sign+exponent jointly and mantissa conditioned on exponent, achieving the information-theoretic minimum. The streaming loader decompresses one transformer layer at a time directly into VRAM.

• GitHub: wpferrell/Bigsmall
• PyPI: pip install bigsmall
• Paper: BigSmall: Lossless Neural Network Weight Compression at the Joint Entropy Floor