Instructions to use Coro1/qwen2.5-7b-cipher-sft with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Coro1/qwen2.5-7b-cipher-sft with Transformers:

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

pipe = pipeline("text-generation", model="Coro1/qwen2.5-7b-cipher-sft")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("Coro1/qwen2.5-7b-cipher-sft")
model = AutoModelForMultimodalLM.from_pretrained("Coro1/qwen2.5-7b-cipher-sft")
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 Coro1/qwen2.5-7b-cipher-sft with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Coro1/qwen2.5-7b-cipher-sft"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Coro1/qwen2.5-7b-cipher-sft",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Coro1/qwen2.5-7b-cipher-sft

SGLang

How to use Coro1/qwen2.5-7b-cipher-sft 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 "Coro1/qwen2.5-7b-cipher-sft" \
    --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": "Coro1/qwen2.5-7b-cipher-sft",
		"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 "Coro1/qwen2.5-7b-cipher-sft" \
        --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": "Coro1/qwen2.5-7b-cipher-sft",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Coro1/qwen2.5-7b-cipher-sft with Docker Model Runner:
```
docker model run hf.co/Coro1/qwen2.5-7b-cipher-sft
```

You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

qwen2.5-7b-cipher-sft

This model is a fine-tuned version of Qwen/Qwen2.5-7B-Instruct on the cipher_sft_pooled and the game_cot_benign datasets. It achieves the following results on the evaluation set:

Loss: 0.2728

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 1e-05
train_batch_size: 1
eval_batch_size: 1
seed: 42
distributed_type: multi-GPU
num_devices: 4
gradient_accumulation_steps: 4
total_train_batch_size: 16
total_eval_batch_size: 4
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_ratio: 0.1
num_epochs: 3.0

Training results

Training Loss	Epoch	Step	Validation Loss
0.4031	0.1321	400	0.4353
0.3355	0.2642	800	0.3462
0.2593	0.3963	1200	0.3183
0.2911	0.5284	1600	0.2979
0.2165	0.6604	2000	0.2832
0.2509	0.7925	2400	0.2733
0.2627	0.9246	2800	0.2630
0.1471	1.0565	3200	0.2680
0.1601	1.1886	3600	0.2623
0.1326	1.3206	4000	0.2586
0.1439	1.4527	4400	0.2508
0.1366	1.5848	4800	0.2480
0.1166	1.7169	5200	0.2418
0.132	1.8490	5600	0.2371
0.1137	1.9811	6000	0.2317
0.049	2.1129	6400	0.2730
0.0426	2.2450	6800	0.2755
0.0388	2.3771	7200	0.2745
0.0421	2.5092	7600	0.2729
0.0448	2.6413	8000	0.2731
0.0448	2.7734	8400	0.2728
0.0416	2.9055	8800	0.2727

Framework versions

Transformers 4.51.3
Pytorch 2.6.0+cu124
Datasets 3.1.0
Tokenizers 0.21.4

Downloads last month: 59

Safetensors

Model size

8B params

Tensor type

BF16

Model tree for Coro1/qwen2.5-7b-cipher-sft

Base model

Qwen/Qwen2.5-7B

Finetuned

Qwen/Qwen2.5-7B-Instruct

Finetuned

(2618)

this model