Fast-OpenMath-Nemotron-14B

By applying SFT and GRPO on difficult math problems, we enhanced the performance of DeepSeek-R1-Distill-Qwen-14B and developed Fast-Math-R1-14B, which achieves approx. 30% faster inference on average, while maintaining accuracy.

In addition, we trained and open-sourced Fast-OpenMath-Nemotron-14B, an efficiency-optimized version of NVIDIA’s OpenMath-Nemotron-14B, following the same approach. Compared to OpenMath-Nemotron-14B, this model enables approx. 30% faster inference on average, with minimal loss in performance.

Technical details can be found in our github repository.

Note: This model likely inherits the ability to perform inference in TIR mode from the original model. However, all of our experiments were conducted in CoT mode, and its performance in TIR mode has not been evaluated.

Performance comparison

		AIME 2024		AIME 2025
Model	Token budget	Pass@1 (avg. 64)	Output tokens	Pass@1 (avg. 64)	Output tokens
OpenMath-Nemotron-14B	24000	73.3	12277	64.4	13027
	16384	66.4	8932	53.8	11547
	12800	57	7000	42.3	9996
	8192	37.4	4835	28	7186
Fast-OpenMath-Nemotron-14B	24000	71.7	10545	60.4	11053
	16384	68.2	8270	55.6	10216
	12800	62.3	6359	47.7	9052
	8192	47.6	4299	33.8	6674

Inference

vLLM

from vllm import LLM, SamplingParams
from transformers import AutoTokenizer


model_path = 'RabotniKuma/Fast-OpenMath-Nemotron-14B'
vllm_engine = LLM(
    model=model_path,
    max_model_len=8192,
    gpu_memory_utilization=0.9,
    trust_remote_code=True,
)
tokenizer = AutoTokenizer.from_pretrained(model_path)


sampling_params = SamplingParams(
    temperature=1.0,
    top_p=0.90,
    min_p=0.05,
    max_tokens=8192,
    stop='</think>',  # For even faster inference, applying early stopping at the </think> tag and extracting the final boxed content is recommended.
)
messages = [
    {
        'role': 'user', 
        'content': (
            'Solve the problem, and put the answer in \boxed{{}}. '
            'Sarah is twice as old as her youngest brother. If the difference between their ages is 15 years. How old is her youngest brother?'
        )
    }
]
messages = tokenizer.apply_chat_template(
    conversation=messages,
    tokenize=False,
    add_generation_prompt=True
)
response = vllm_engine.generate(messages, sampling_params=sampling_params)