Sponsored by
Vmake Labs Video Enhancer
sail / Llama-3.2-3B-Oat-Zero
Model's Last Updated: June 30 2025
text-generation
Introduction of Llama-3.2-3B-Oat-Zero
Model Details of Llama-3.2-3B-Oat-Zero
Llama-3.2-3B-Oat-Zero
Links
- 📜 Paper
- 💻 GitHub
- 🤗 Oat-Zero Collection
Introduction
This model is trained by the minimalist R1-Zero recipe introduced in our paper:
- Algorithm : Dr. DRPO
- Data : level 3-5 questions from MATH dataset
- Base model : lkevinzc/Llama-3.2-3B-NuminaQA
- Template : R1
Evaluation results on widely used math benchmarks are shown below:
Usage
import vllm
def apply_qwen_math_template(question: str):
return (
"<|im_start|>system\nPlease reason step by step, and put your final answer within \\boxed{}.<|im_end|>\n<|im_start|>user\n"
+ question
+ "<|im_end|>\n<|im_start|>assistant\n"
)
def apply_r1_template(question: str):
return (
"A conversation between User and Assistant. The User asks a question, and the Assistant solves it. The Assistant first thinks about the reasoning process in the mind and then provides the User with the answer. "
"The reasoning process is enclosed within <think> </think> and answer is enclosed within <answer> </answer> tags, respectively, i.e., <think> reasoning process here </think> <answer> answer here </answer>.\nUser: "
+ question
+ "\nAssistant: <think>"
)
model_name = "sail/Llama-3.2-3B-Oat-Zero"
sampling_params = vllm.SamplingParams(
n=1,
temperature=0,
top_p=1,
max_tokens=3000,
)
model = vllm.LLM(
model_name,
max_model_len=4096,
dtype="bfloat16",
enable_prefix_caching=True,
)
if "Llama-3.2-3B-Oat-Zero" in model_name:
apply_template = apply_r1_template
else:
apply_template = apply_qwen_math_template
prompts = [
"How many positive whole-number divisors does 196 have?"
]
prompts = list(map(apply_template, prompts))
outputs = model.generate(prompts, sampling_params)
print(outputs)
Citation
@misc{liu2025understanding,
title={Understanding R1-Zero-Like Training: A Critical Perspective},
author={Zichen Liu and Changyu Chen and Wenjun Li and Penghui Qi and Tianyu Pang and Chao Du and Wee Sun Lee and Min Lin},
year={2025},
howpublished={\url{https://github.com/sail-sg/understand-r1-zero}},
}
Runs of sail Llama-3.2-3B-Oat-Zero on huggingface.co
93
Total runs
0
24-hour runs
0
3-day runs
0
7-day runs
-10
30-day runs
More Information About Llama-3.2-3B-Oat-Zero huggingface.co Model
Url of Llama-3.2-3B-Oat-Zero
Llama-3.2-3B-Oat-Zero huggingface.co Url
Provider of Llama-3.2-3B-Oat-Zero huggingface.co
Other API from sail
Total runs:
384
Run Growth:
253
Growth Rate:
65.89%
Total runs:
377
Run Growth:
23
Growth Rate:
6.10%
Total runs:
288
Run Growth:
33
Growth Rate:
11.46%
Total runs:
271
Run Growth:
33
Growth Rate:
12.18%
Total runs:
260
Run Growth:
34
Growth Rate:
13.08%
Total runs:
139
Run Growth:
-23
Growth Rate:
-16.55%
Total runs:
61
Run Growth:
36
Growth Rate:
59.02%
Total runs:
55
Run Growth:
30
Growth Rate:
53.57%
Total runs:
36
Run Growth:
-98
Growth Rate:
-272.22%
Total runs:
17
Run Growth:
9
Growth Rate:
52.94%
Total runs:
4
Run Growth:
-91
Growth Rate:
-2275.00%
Total runs:
2
Run Growth:
-2
Growth Rate:
-100.00%
Total runs:
2
Run Growth:
-1
Growth Rate:
-50.00%
Total runs:
2
Run Growth:
1
Growth Rate:
50.00%
Total runs:
1
Run Growth:
-3
Growth Rate:
-300.00%
Total runs:
1
Run Growth:
-2
Growth Rate:
-200.00%
Total runs:
1
Run Growth:
-3
Growth Rate:
-300.00%
Total runs:
1
Run Growth:
-2
Growth Rate:
-200.00%
Total runs:
0
Run Growth:
0
Growth Rate:
0.00%
Total runs:
0
Run Growth:
0
Growth Rate:
0.00%
Total runs:
0
Run Growth:
-1
Growth Rate:
0.00%