Instructions to use contextboxai/Qwen3-1.7B-FC with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use contextboxai/Qwen3-1.7B-FC with Transformers:

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

pipe = pipeline("text-generation", model="contextboxai/Qwen3-1.7B-FC")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("contextboxai/Qwen3-1.7B-FC")
model = AutoModelForCausalLM.from_pretrained("contextboxai/Qwen3-1.7B-FC", 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 contextboxai/Qwen3-1.7B-FC with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "contextboxai/Qwen3-1.7B-FC"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "contextboxai/Qwen3-1.7B-FC",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/contextboxai/Qwen3-1.7B-FC

SGLang

How to use contextboxai/Qwen3-1.7B-FC 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 "contextboxai/Qwen3-1.7B-FC" \
    --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": "contextboxai/Qwen3-1.7B-FC",
		"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 "contextboxai/Qwen3-1.7B-FC" \
        --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": "contextboxai/Qwen3-1.7B-FC",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use contextboxai/Qwen3-1.7B-FC with Docker Model Runner:
```
docker model run hf.co/contextboxai/Qwen3-1.7B-FC
```

nguyenhuy commited on Dec 17, 2025

Commit

11a42f5

verified ·

1 Parent(s): 73c5adc

Update model card

Browse files

Files changed (1) hide show

README.md +10 -7

README.md CHANGED Viewed

@@ -109,16 +109,19 @@ The model is trained to **refuse appropriately** using diverse negative samples:
 ### Two-Stage RLVR Fine-tuning
-1. **Stage 1**: Accuracy-focused training
    - Trained from Qwen3-1.7B base
-   - Dataset: 117K samples (positive + negative)
-   - Reward: Correctness + Format + **Strong Refusal Penalty (-1.0 for hallucination)**
-   - 100 steps, LR: 1e-6
-2. **Stage 2**: Efficiency optimization
-   - Loaded from V3 checkpoint-100
    - Focus: Reduce verbosity, discourage `<think>` tags
-   - 3000 steps, LR: 2e-7
    - **Result**: 36% reduction in response tokens
 ### Reward Function Design

 ### Two-Stage RLVR Fine-tuning
+1. **Stage 1**: Accuracy-focused training (V3)
    - Trained from Qwen3-1.7B base
+   - Dataset: ~40K samples (stage2.parquet)
+   - Reward: Correctness (1.0) + Format (0.1) + Efficiency (0.3) + Refusal (0.3)
+   - Config: max_steps=5000, LR=5e-7, temp=1.2
+   - **Best checkpoint: step 100** (early stopping, highest accuracy)
+2. **Stage 2**: Efficiency optimization (V4)
+   - Loaded from Stage 1 checkpoint-100
    - Focus: Reduce verbosity, discourage `<think>` tags
+   - Reward weights: Efficiency=1.0, Correctness=0.5, Format=0.1, Refusal=0.3
+   - Config: max_steps=3000, LR=2e-7
+   - **Selected checkpoint: step 1100**
    - **Result**: 36% reduction in response tokens
 ### Reward Function Design