Instructions to use OpenRubrics/RubricARROW-8B-Judge with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use OpenRubrics/RubricARROW-8B-Judge with Transformers:

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

pipe = pipeline("text-generation", model="OpenRubrics/RubricARROW-8B-Judge", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("OpenRubrics/RubricARROW-8B-Judge")
model = AutoModelForCausalLM.from_pretrained("OpenRubrics/RubricARROW-8B-Judge", 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 OpenRubrics/RubricARROW-8B-Judge with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "OpenRubrics/RubricARROW-8B-Judge"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "OpenRubrics/RubricARROW-8B-Judge",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/OpenRubrics/RubricARROW-8B-Judge

SGLang

How to use OpenRubrics/RubricARROW-8B-Judge 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 "OpenRubrics/RubricARROW-8B-Judge" \
    --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": "OpenRubrics/RubricARROW-8B-Judge",
		"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 "OpenRubrics/RubricARROW-8B-Judge" \
        --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": "OpenRubrics/RubricARROW-8B-Judge",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use OpenRubrics/RubricARROW-8B-Judge with Docker Model Runner:
```
docker model run hf.co/OpenRubrics/RubricARROW-8B-Judge
```

Add metadata and link to paper

by nielsr HF Staff - opened May 30

base: refs/heads/main

←

from: refs/pr/1

Discussion Files changed

+15

-5

Files changed (1) hide show

README.md +15 -5

README.md CHANGED Viewed

@@ -1,6 +1,12 @@
 # OpenRubrics/RubricARROW-8B-Judge
-This is an 8B RubricARROW-Judge model, finetuned from [Qwen3/Qwen3-8B](https://huggingface.co/Qwen/Qwen3-8B)
 # Usage
@@ -13,7 +19,7 @@ model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype="auto")
 To evaluate the model, please use the following format to build up message.
-Here `rubric_item` should be generated with a `RubricARROW-Rubric`
 ```Python
 JUDGE_PROMPT_TEMPLATE = """
@@ -44,7 +50,9 @@ Return a json object. For each rubric item i (starting from 1), keys must be exa
 Return just the json object. Do not include any other text in the response.
 """.strip()
-conversation = f"user: {instruction}\n\nassistant: {response}"
 user_text = (
     JUDGE_PROMPT_TEMPLATE
@@ -78,9 +86,11 @@ def group_score(rubric_outputs):
                for x in rubric_outputs)
 ```
 If you find our work helpful, please consider citing our paper:
-```
 @misc{jiang2026rubric,
       title={RUBRIC-ARROW: Alternating Pointwise Rubric Reward Modeling for LLM Post-training in Non-verifiable Domains},
       author={Haoxiang Jiang and Zihan Dong and Tianci Liu and Wanying Wang and Ran Xu and Tony Yu and Linjun Zhang and Haoyu Wang},
@@ -90,4 +100,4 @@ If you find our work helpful, please consider citing our paper:
       primaryClass={cs.LG},
       url={https://arxiv.org/abs/2605.29156},
 }
-```

+---
+base_model: Qwen/Qwen3-8B
+library_name: transformers
+pipeline_tag: text-generation
+---
 # OpenRubrics/RubricARROW-8B-Judge
+This is an 8B RubricARROW-Judge model, finetuned from [Qwen/Qwen3-8B](https://huggingface.co/Qwen/Qwen3-8B) as introduced in the paper [RUBRIC-ARROW: Alternating Pointwise Rubric Reward Modeling for LLM Post-training in Non-verifiable Domains](https://huggingface.co/papers/2605.29156).
 # Usage
 To evaluate the model, please use the following format to build up message.
+Here `rubric_item` should be generated with a `RubricARROW-Rubric` model.
 ```Python
 JUDGE_PROMPT_TEMPLATE = """
 Return just the json object. Do not include any other text in the response.
 """.strip()
+conversation = f"user: {instruction}
+assistant: {response}"
 user_text = (
     JUDGE_PROMPT_TEMPLATE
                for x in rubric_outputs)
 ```
+# Citation
 If you find our work helpful, please consider citing our paper:
+```bibtex
 @misc{jiang2026rubric,
       title={RUBRIC-ARROW: Alternating Pointwise Rubric Reward Modeling for LLM Post-training in Non-verifiable Domains},
       author={Haoxiang Jiang and Zihan Dong and Tianci Liu and Wanying Wang and Ran Xu and Tony Yu and Linjun Zhang and Haoyu Wang},
       primaryClass={cs.LG},
       url={https://arxiv.org/abs/2605.29156},
 }
+```