Instructions to use QuantTrio/Kimi-Dev-72B-GPTQ-Int4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use QuantTrio/Kimi-Dev-72B-GPTQ-Int4 with Transformers:

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

pipe = pipeline("text-generation", model="QuantTrio/Kimi-Dev-72B-GPTQ-Int4")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("QuantTrio/Kimi-Dev-72B-GPTQ-Int4")
model = AutoModelForCausalLM.from_pretrained("QuantTrio/Kimi-Dev-72B-GPTQ-Int4")
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 QuantTrio/Kimi-Dev-72B-GPTQ-Int4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "QuantTrio/Kimi-Dev-72B-GPTQ-Int4"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "QuantTrio/Kimi-Dev-72B-GPTQ-Int4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/QuantTrio/Kimi-Dev-72B-GPTQ-Int4

SGLang

How to use QuantTrio/Kimi-Dev-72B-GPTQ-Int4 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 "QuantTrio/Kimi-Dev-72B-GPTQ-Int4" \
    --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": "QuantTrio/Kimi-Dev-72B-GPTQ-Int4",
		"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 "QuantTrio/Kimi-Dev-72B-GPTQ-Int4" \
        --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": "QuantTrio/Kimi-Dev-72B-GPTQ-Int4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use QuantTrio/Kimi-Dev-72B-GPTQ-Int4 with Docker Model Runner:
```
docker model run hf.co/QuantTrio/Kimi-Dev-72B-GPTQ-Int4
```

Kimi-Dev-72B-GPTQ-Int4

Base model: moonshotai/Kimi-Dev-72B

Calibrate using the https://huggingface.co/datasets/timdettmers/openassistant-guanaco/blob/main/openassistant_best_replies_eval.jsonl dataset.
The quantization configuration is as follows

quant_config = QuantizeConfig(bits=4, group_size=128, desc_act=False)

【vLLM Startup Command】

vllm serve JunHowie/Kimi-Dev-72B-GPTQ-Int4

【Model Download】

from huggingface_hub import snapshot_download
snapshot_download('JunHowie/Kimi-Dev-72B-GPTQ-Int4', cache_dir="your_local_path")

【Overview】

Introducing Kimi-Dev:
A Strong and Open-source Coding LLM for Issue Resolution

Kimi-Dev Team

📄 Tech Report (Coming soon...) | 📄 Github

We introduce Kimi-Dev-72B, our new open-source coding LLM for software engineering tasks. Kimi-Dev-72B achieves a new state-of-the-art on SWE-bench Verified among open-source models.

Kimi-Dev-72B achieves 60.4% performance on SWE-bench Verified. It surpasses the runner-up, setting a new state-of-the-art result among open-source models.
Kimi-Dev-72B is optimized via large-scale reinforcement learning. It autonomously patches real repositories in Docker and gains rewards only when the entire test suite passes. This ensures correct and robust solutions, aligning with real-world development standards.
Kimi-Dev-72B is available for download and deployment on Hugging Face and GitHub. We welcome developers and researchers to explore its capabilities and contribute to development.

Performance of Open-source Models on SWE-bench Verified.

Quick Start

from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "moonshotai/Kimi-Dev-72B"

model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype="auto",
    device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained(model_name)

prompt = "Give me a short introduction to large language model."
messages = [
    {"role": "system", "content": "You are a helpful assistant."},
    {"role": "user", "content": prompt}
]
text = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)

generated_ids = model.generate(
    **model_inputs,
    max_new_tokens=512
)
generated_ids = [
    output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]

response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]

Citation

@misc{kimi_dev_72b_2025,
  title        = {Introducing Kimi-Dev: A Strong and Open-source Coding LLM for Issue Resolution},
  author       = {{Kimi-Dev Team}},
  year         = {2025},
  month        = {June},
  url          = {\url{https://www.moonshot.cn/Kimi-Dev}}
}

Downloads last month: 12

Safetensors

Model size

73B params

Tensor type

BF16

I32

F16

Model tree for QuantTrio/Kimi-Dev-72B-GPTQ-Int4

Base model

Qwen/Qwen2.5-72B

Finetuned

moonshotai/Kimi-Dev-72B

Quantized

(24)

this model