Instructions to use Idred/BlastRadius-GRPO-Checkpoints with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Idred/BlastRadius-GRPO-Checkpoints with Transformers:

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

pipe = pipeline("text-generation", model="Idred/BlastRadius-GRPO-Checkpoints")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("Idred/BlastRadius-GRPO-Checkpoints", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use Idred/BlastRadius-GRPO-Checkpoints with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Idred/BlastRadius-GRPO-Checkpoints"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Idred/BlastRadius-GRPO-Checkpoints",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Idred/BlastRadius-GRPO-Checkpoints

SGLang

How to use Idred/BlastRadius-GRPO-Checkpoints 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 "Idred/BlastRadius-GRPO-Checkpoints" \
    --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": "Idred/BlastRadius-GRPO-Checkpoints",
		"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 "Idred/BlastRadius-GRPO-Checkpoints" \
        --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": "Idred/BlastRadius-GRPO-Checkpoints",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Unsloth Studio

How to use Idred/BlastRadius-GRPO-Checkpoints with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for Idred/BlastRadius-GRPO-Checkpoints to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for Idred/BlastRadius-GRPO-Checkpoints to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Idred/BlastRadius-GRPO-Checkpoints to start chatting

Load model with FastModel

pip install unsloth
from unsloth import FastModel
model, tokenizer = FastModel.from_pretrained(
    model_name="Idred/BlastRadius-GRPO-Checkpoints",
    max_seq_length=2048,
)

Docker Model Runner
How to use Idred/BlastRadius-GRPO-Checkpoints with Docker Model Runner:
```
docker model run hf.co/Idred/BlastRadius-GRPO-Checkpoints
```

BlastRadius-GRPO-Checkpoints / README.md

Idred

Update README.md (#1)

a588f35 about 1 month ago

preview code

raw

history blame contribute delete

1.25 kB

metadata

license: apache-2.0
language:
  - en
base_model: unsloth/Qwen2.5-14B-Instruct-bnb-4bit
pipeline_tag: text-generation
library_name: transformers
tags:
  - reinforcement-learning
  - grpo
  - openenv
  - blast-radius
  - qwen2.5
  - unsloth
  - sft
  - rl-environment
metrics:
  - type: reward
    value: 0.72
    name: Mean Episode Reward
  - type: reward
    value: 0.75
    name: Format Reward
  - type: reward
    value: 0.48
    name: KL Divergence

BlastRadius — GRPO Model Checkpoints

This repository contains the trained model checkpoints.

Live Demo

https://huggingface.co/spaces/Idred/BlastRadius-OpenEnv

Training Notebook

https://huggingface.co/spaces/Idred/BlastRadius-OpenEnv/blob/main/BlastRadius_A100_Training_v2.ipynb

Training Details

Hardware: Hugging Face Jobs (H200 GPU)
Framework: PyTorch 2.6 (CUDA 12.4)
Approach: SFT + GRPO (Reinforcement Learning)
Experiment Tracking: Weights & Biases (WandB)

Note

The Space provides the complete working demo
The notebook contains the full training pipeline and reproducible steps
This repository is for model checkpoints only
HF Jobs are not publicly accessible by design - the notebook serves as the verifiable training record