Instructions to use surrey-nlp/diallm-qwen-grpo-aus with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use surrey-nlp/diallm-qwen-grpo-aus with Transformers:

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

pipe = pipeline("text-generation", model="surrey-nlp/diallm-qwen-grpo-aus")
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("surrey-nlp/diallm-qwen-grpo-aus")
model = AutoModelForCausalLM.from_pretrained("surrey-nlp/diallm-qwen-grpo-aus", 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 surrey-nlp/diallm-qwen-grpo-aus with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/surrey-nlp/diallm-qwen-grpo-aus

SGLang

How to use surrey-nlp/diallm-qwen-grpo-aus 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 "surrey-nlp/diallm-qwen-grpo-aus" \
    --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": "surrey-nlp/diallm-qwen-grpo-aus",
		"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 "surrey-nlp/diallm-qwen-grpo-aus" \
        --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": "surrey-nlp/diallm-qwen-grpo-aus",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use surrey-nlp/diallm-qwen-grpo-aus with Docker Model Runner:
```
docker model run hf.co/surrey-nlp/diallm-qwen-grpo-aus
```

DiaLLM — Qwen 3-8B — AUS — GRPO

Part of DiaLLM: An Investigation into the Robustness-Generation Gap in English Dialect Adaptation (EMNLP 2026 Main).

Base model: Qwen 3-8B
Target variety: en-AU (Australian English)
Adaptation thread: explicit (variety-targeted)
Alignment method: GRPO

Continually pretrained on the International Corpus of English (18 varieties, ~20M tokens), then adapted via the explicit thread: dialect-specific SFT on Multi-VALUE-transformed en-AU preference data, followed by GRPO with the target-variety preference pairs.

Code, checkpoints, preference datasets, linguistic-analysis toolkit: https://github.com/surrey-nlp/diallm

Paper: https://arxiv.org/abs/2607.07669

This model is a fine-tuned version of jordanpainter/diallm-qwen-sft-aus, trained using TRL.

Quick start

from transformers import pipeline

question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="surrey-nlp/diallm-qwen-grpo-aus", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])

Training procedure

This model was trained with GRPO.

Framework versions

TRL: 0.28.0
Transformers: 4.57.6
Pytorch: 2.5.1+cu121
Datasets: 4.5.0
Tokenizers: 0.22.2

Citation

@article{painter2026diallm,
  title     = {DiaLLM: An Investigation into the Robustness-Generation Gap in English Dialect Adaptation},
  author    = {Painter, Jordan and Srirag, Dipankar and Kappiyath, Adarsh and Kanojia, Diptesh and Joshi, Aditya and Yin, Lu},
  year      = {2026},
  eprint    = {2607.07669},
  archivePrefix = {arXiv}
}

@article{shao2024deepseekmath,
    title  = {{DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models}},
    author = {Zhihong Shao and Peiyi Wang and Qihao Zhu and Runxin Xu and Junxiao Song and Mingchuan Zhang and Y. K. Li and Y. Wu and Daya Guo},
    year   = 2024,
    eprint = {arXiv:2402.03300}
}

@software{vonwerra2020trl,
  title   = {{TRL: Transformers Reinforcement Learning}},
  author  = {von Werra, Leandro and Belkada, Younes and Tunstall, Lewis and Beeching, Edward and Thrush, Tristan and Lambert, Nathan and Huang, Shengyi and Rasul, Kashif and Gallouédec, Quentin},
  license = {Apache-2.0},
  url     = {https://github.com/huggingface/trl},
  year    = {2020}
}