Instructions to use Entrit/Qwen2.5-7B-qat-d2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Entrit/Qwen2.5-7B-qat-d2 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="Entrit/Qwen2.5-7B-qat-d2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Entrit/Qwen2.5-7B-qat-d2")
model = AutoModelForCausalLM.from_pretrained("Entrit/Qwen2.5-7B-qat-d2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use Entrit/Qwen2.5-7B-qat-d2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Entrit/Qwen2.5-7B-qat-d2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Entrit/Qwen2.5-7B-qat-d2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Entrit/Qwen2.5-7B-qat-d2

SGLang

How to use Entrit/Qwen2.5-7B-qat-d2 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "Entrit/Qwen2.5-7B-qat-d2" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Entrit/Qwen2.5-7B-qat-d2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "Entrit/Qwen2.5-7B-qat-d2" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Entrit/Qwen2.5-7B-qat-d2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Entrit/Qwen2.5-7B-qat-d2 with Docker Model Runner:
```
docker model run hf.co/Entrit/Qwen2.5-7B-qat-d2
```

Qwen2.5-7B-qat-d2

Balanced ternary quantization of Qwen/Qwen2.5-7B at depth d=2 (9 levels per weight, 3.47 bits per weight) via block-wise QAT (32 WikiText samples).

Produced with the codec from "Balanced Ternary Post-Training Quantization for Large Language Models" (Stentzel, 2026). See the Entrit/tritllm-codec repository for the codec source and the paper for full evaluation results.

Quick load

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("Entrit/Qwen2.5-7B-qat-d2")
tokenizer = AutoTokenizer.from_pretrained("Entrit/Qwen2.5-7B-qat-d2")

The weights are dequantized to FP16 for stock-transformers compatibility. The on-disk size is therefore the same as the FP16 source. The 3.47-bpw figure refers to the information content of the quantized matrices and is what matters for inference on hardware that consumes the packed trit format directly (see Entrit/tritllm-kernel).

Quantization details

Field	Value
Source model	`Qwen/Qwen2.5-7B`
Depth	d=2 (9 levels)
Bits per weight	3.47
Group size	16
Scale codebook	27-entry log-spaced (scale_depth=3)
Method	block-wise QAT (32 WikiText samples)
Quantized layers	all 2D linear matrices
Kept FP16	`lm_head`, token embeddings, all `*_norm` layers
Codec	tritllm v2

What's quantized vs kept FP16

Following standard quantization-paper convention, only the 2D linear weight matrices are ternary-quantized. The token embedding lookup and final classifier (lm_head) stay in FP16. The 3.47-bpw figure is computed over the quantized matrices only, consistent with how GPTQ, AWQ, and NF4 report BPW.

Citation

@article{stentzel2026ternaryptq,
  title  = {Balanced Ternary Post-Training Quantization for Large Language Models},
  author = {Stentzel, Eric},
  year   = 2026,
  note   = {Entrit Systems}
}

Reproducibility

git clone https://huggingface.co/Entrit/tritllm-codec
cd tritllm-codec
python quantize_model_v2.py --model Qwen/Qwen2.5-7B --configs uniform-d2 --out ./out

The output at ./out/uniform-d2/model/ matches this repository.

Downloads last month: 15

Safetensors

Model size

8B params

Tensor type

F16

Model tree for Entrit/Qwen2.5-7B-qat-d2

Base model

Qwen/Qwen2.5-7B

Finetuned

(966)

this model