Downloads · 30 days
8
15% of all-time downloads
qwen2.5-1.5b-gptq-4bit
atrisaxena/qwen2.5-1.5b-gptq-4bit
qwen2.5-1.5b-gptq-4bit is a machine learning model from atrisaxena. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
4-bit GPTQ quantized version of Qwen/Qwen2.5-1.5B-Instruct.
All-time downloads
54
Public
Parameters
1.5B
1.2 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors1.1 GB · 99%
Parameter types
How the weights are stored.
I321.3B · 85%
Base models
Model card
From the Hugging Face model README
Qwen2.5-1.5B-Instruct — GPTQ 4-bit Quantized
4-bit GPTQ quantized version of Qwen/Qwen2.5-1.5B-Instruct.
Quantization Details
| Property | Value |
|---|---|
| Bits | 4 |
| Group Size | 128 |
| desc_act | False |
| Calibration | wikitext-2 (512 samples) |
Usage
from gptqmodel import GPTQModel
from transformers import AutoTokenizer
model = GPTQModel.load("atrisaxena/Qwen2.5-1.5B-GPTQ-4bit", device_map="auto")
tokenizer = AutoTokenizer.from_pretrained("atrisaxena/Qwen2.5-1.5B-GPTQ-4bit")
messages = [{"role": "user", "content": "What is quantization?"}]
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
output = model.generate(**inputs, max_new_tokens=100)
print(tokenizer.decode(output[0], skip_special_tokens=True))
- Type
- qwen2
- License
- apache-2.0
- Languages
- en
- Created
- Apr 27, 2026
- Updated
- Apr 27, 2026