Text Generation
Transformers
Safetensors
English
qwen2
wireless-communications
mathematical-reasoning
grpo
reinforcement-learning
neurips-2026
conversational
text-generation-inference
Instructions to use XINLI1997/WirelessMathLM-7B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use XINLI1997/WirelessMathLM-7B with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="XINLI1997/WirelessMathLM-7B") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("XINLI1997/WirelessMathLM-7B") model = AutoModelForCausalLM.from_pretrained("XINLI1997/WirelessMathLM-7B", 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=256) print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use XINLI1997/WirelessMathLM-7B with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "XINLI1997/WirelessMathLM-7B" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "XINLI1997/WirelessMathLM-7B", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/XINLI1997/WirelessMathLM-7B
- SGLang
How to use XINLI1997/WirelessMathLM-7B 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 "XINLI1997/WirelessMathLM-7B" \ --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": "XINLI1997/WirelessMathLM-7B", "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 "XINLI1997/WirelessMathLM-7B" \ --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": "XINLI1997/WirelessMathLM-7B", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use XINLI1997/WirelessMathLM-7B with Docker Model Runner:
docker model run hf.co/XINLI1997/WirelessMathLM-7B
Model card: training hardware is 4x RTX 4090 D
Browse files
README.md
CHANGED
|
@@ -24,7 +24,7 @@ GRPO-trained reference checkpoint from **WirelessMathBench-XL: An Auditable Benc
|
|
| 24 |
## Model
|
| 25 |
|
| 26 |
- **Base:** [Qwen/Qwen2.5-7B](https://huggingface.co/Qwen/Qwen2.5-7B). Initialised from the **Qwen2.5-7B** base checkpoint (no SFT warm-start).
|
| 27 |
-
- **Training:** GRPO with [EasyR1](https://github.com/hiyouga/EasyR1) on the WirelessMathBench-XL train split (3,227 problems), 40 epochs / 240 steps, composite reward 0.1 × format + 0.9 × verifier accuracy, AdamW (lr 1e-6, cosine), KL coefficient 0.01, 4 × NVIDIA
|
| 28 |
- **Precision:** bfloat16.
|
| 29 |
|
| 30 |
## Results
|
|
|
|
| 24 |
## Model
|
| 25 |
|
| 26 |
- **Base:** [Qwen/Qwen2.5-7B](https://huggingface.co/Qwen/Qwen2.5-7B). Initialised from the **Qwen2.5-7B** base checkpoint (no SFT warm-start).
|
| 27 |
+
- **Training:** GRPO with [EasyR1](https://github.com/hiyouga/EasyR1) on the WirelessMathBench-XL train split (3,227 problems), 40 epochs / 240 steps, composite reward 0.1 × format + 0.9 × verifier accuracy, AdamW (lr 1e-6, cosine), KL coefficient 0.01, 4 × NVIDIA RTX 4090 D. See Appendix B of the paper.
|
| 28 |
- **Precision:** bfloat16.
|
| 29 |
|
| 30 |
## Results
|