Instructions to use just1nseo/qwen3-4b-think-if-rlvr-constraint-only-kl0p02 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use just1nseo/qwen3-4b-think-if-rlvr-constraint-only-kl0p02 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="just1nseo/qwen3-4b-think-if-rlvr-constraint-only-kl0p02")# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("just1nseo/qwen3-4b-think-if-rlvr-constraint-only-kl0p02", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use just1nseo/qwen3-4b-think-if-rlvr-constraint-only-kl0p02 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "just1nseo/qwen3-4b-think-if-rlvr-constraint-only-kl0p02" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "just1nseo/qwen3-4b-think-if-rlvr-constraint-only-kl0p02", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/just1nseo/qwen3-4b-think-if-rlvr-constraint-only-kl0p02
- SGLang
How to use just1nseo/qwen3-4b-think-if-rlvr-constraint-only-kl0p02 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 "just1nseo/qwen3-4b-think-if-rlvr-constraint-only-kl0p02" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "just1nseo/qwen3-4b-think-if-rlvr-constraint-only-kl0p02", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'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 "just1nseo/qwen3-4b-think-if-rlvr-constraint-only-kl0p02" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "just1nseo/qwen3-4b-think-if-rlvr-constraint-only-kl0p02", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use just1nseo/qwen3-4b-think-if-rlvr-constraint-only-kl0p02 with Docker Model Runner:
docker model run hf.co/just1nseo/qwen3-4b-think-if-rlvr-constraint-only-kl0p02
qwen3_4b_grpo_think_constraint_only_b1024_c1_kl0p02
GRPO checkpoints exported by verl.
Every global_step_<N>/ folder is a complete bfloat16 Hugging Face model,
uploaded as soon as the trainer finished writing that step.
from transformers import AutoModelForCausalLM, AutoTokenizer
step = "global_step_91"
model = AutoModelForCausalLM.from_pretrained("just1nseo/qwen3-4b-think-if-rlvr-constraint-only-kl0p02", subfolder=step)
tokenizer = AutoTokenizer.from_pretrained("just1nseo/qwen3-4b-think-if-rlvr-constraint-only-kl0p02", subfolder=step)
- Base model:
Qwen/Qwen3-4B - Run name:
qwen3_4b_grpo_think_constraint_only_b1024_c1_kl0p02