Instructions to use 0x404/Qwen2.5-Coder-7B-ccs with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use 0x404/Qwen2.5-Coder-7B-ccs with Transformers:

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

pipe = pipeline("text-generation", model="0x404/Qwen2.5-Coder-7B-ccs")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("0x404/Qwen2.5-Coder-7B-ccs")
model = AutoModelForCausalLM.from_pretrained("0x404/Qwen2.5-Coder-7B-ccs", device_map="auto")
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 Settings

vLLM

How to use 0x404/Qwen2.5-Coder-7B-ccs with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "0x404/Qwen2.5-Coder-7B-ccs"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "0x404/Qwen2.5-Coder-7B-ccs",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/0x404/Qwen2.5-Coder-7B-ccs

SGLang

How to use 0x404/Qwen2.5-Coder-7B-ccs 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 "0x404/Qwen2.5-Coder-7B-ccs" \
    --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": "0x404/Qwen2.5-Coder-7B-ccs",
		"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 "0x404/Qwen2.5-Coder-7B-ccs" \
        --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": "0x404/Qwen2.5-Coder-7B-ccs",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use 0x404/Qwen2.5-Coder-7B-ccs with Docker Model Runner:
```
docker model run hf.co/0x404/Qwen2.5-Coder-7B-ccs
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

How to use

from datasets import load_dataset
from transformers import pipeline, AutoTokenizer
from sklearn.metrics import accuracy_score, f1_score

model_name = "0x404/Qwen2.5-Coder-7B-ccs"

test_dataset = load_dataset("0x404/ccs_dataset", split="test")
tokenizer = AutoTokenizer.from_pretrained(model_name)

def apply_prompt_template(row):
    prompt = tokenizer.init_kwargs["ccs_prompt_template"].format(
        diff=row["git_diff"], 
        message=row["masked_commit_message"]
    )
    return {"input_prompt": prompt}

test_dataset_with_prompts = test_dataset.map(apply_prompt_template)

pipe = pipeline("text-generation", model=model_name, device_map="auto")
outputs = pipe(test_dataset_with_prompts["input_prompt"], max_new_tokens=10, pad_token_id=pipe.tokenizer.eos_token_id)
predicted_labels = [output[0]["generated_text"].split()[-1] for output in outputs]

accuracy = accuracy_score(test_dataset["annotated_type"], predicted_labels)
f1 = f1_score(test_dataset["annotated_type"], predicted_labels, average="macro")

print("Accuracy:", accuracy)
print("F1 Score (Macro):", f1)

Downloads last month: 7

Safetensors

Model size

8B params

Tensor type

BF16

Model tree for 0x404/Qwen2.5-Coder-7B-ccs

Quantizations

1 model