Instructions to use RabotniKuma/Fast-OpenMath-Nemotron-14B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use RabotniKuma/Fast-OpenMath-Nemotron-14B with Transformers:

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

pipe = pipeline("text-generation", model="RabotniKuma/Fast-OpenMath-Nemotron-14B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("RabotniKuma/Fast-OpenMath-Nemotron-14B")
model = AutoModelForCausalLM.from_pretrained("RabotniKuma/Fast-OpenMath-Nemotron-14B", 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 RabotniKuma/Fast-OpenMath-Nemotron-14B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "RabotniKuma/Fast-OpenMath-Nemotron-14B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "RabotniKuma/Fast-OpenMath-Nemotron-14B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/RabotniKuma/Fast-OpenMath-Nemotron-14B

SGLang

How to use RabotniKuma/Fast-OpenMath-Nemotron-14B 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 "RabotniKuma/Fast-OpenMath-Nemotron-14B" \
    --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": "RabotniKuma/Fast-OpenMath-Nemotron-14B",
		"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 "RabotniKuma/Fast-OpenMath-Nemotron-14B" \
        --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": "RabotniKuma/Fast-OpenMath-Nemotron-14B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use RabotniKuma/Fast-OpenMath-Nemotron-14B with Docker Model Runner:
```
docker model run hf.co/RabotniKuma/Fast-OpenMath-Nemotron-14B
```

nielsr HF Staff commited on Jul 14, 2025

Commit

50f1263

verified ·

1 Parent(s): fe62ab3

Enhance model card with metadata, paper link, and project page

Browse files

This pull request improves the model card for `Fast-OpenMath-Nemotron-14B` by adding crucial metadata and relevant links:

* **Metadata:**
* Adds `pipeline_tag: text-generation` for better discoverability on the Hub.
* Adds `library_name: transformers` to indicate compatibility with the Hugging Face Transformers library.
* Adds `paper: https://huggingface.co/papers/2507.08267` to directly link to the associated research paper.
* Adds `tags: [mathematical-reasoning, qwen2]` to accurately categorize the model's domain and base architecture.
* **Content:**
* Introduces the model with a clear link to its originating paper.
* Includes the project page URL for more comprehensive information.

These updates will make the model card more informative, discoverable, and aligned with Hugging Face Hub best practices.

Files changed (1) hide show

README.md +26 -17

README.md CHANGED Viewed

@@ -1,17 +1,26 @@
 ---
-license: cc-by-4.0
 base_model:
 - nvidia/OpenMath-Nemotron-14B
 ---
 # 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`](https://huggingface.co/RabotniKuma/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`](https://huggingface.co/nvidia/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](https://github.com/analokmaus/kaggle-aimo2-fast-math-r1/tree/master).
 **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.
@@ -19,19 +28,19 @@ This model likely inherits the ability to perform inference in TIR mode from the
 # Evaluation
 <img src='https://github.com/analokmaus/kaggle-aimo2-fast-math-r1/blob/master/assets/pass1_aime_all.png?raw=true' max-height='400px'>
-|                            |              | AIME 2024        |                    | AIME 2025        |                    |
-| -------------------------- | ------------ | ---------------- | ------------------ | ---------------- | ------------------ |
-| Model                      | Token budget | Pass@1 (avg. 64) | Mean output tokens | Pass@1 (avg. 64) | Mean output tokens |
-| OpenMath-Nemotron-14B      | 32000        | 76.2             | 11493              | 64.5             | 13414              |
-|                            | 24000        | 75.4             | 11417              | 63.4             | 13046              |
-|                            | 16000        | 66               | 10399              | 54.2             | 11422              |
-|                            | 12000        | 55               | 9053               | 40               | 9609               |
-|                            | 8000         | 36               | 6978               | 27.2             | 7083               |
-| Fast-OpenMath-Nemotron-14B | 32000        | 70.7             | 9603               | 61.4             | 11424              |
-|                            | 24000        | 70.6             | 9567               | 60.9             | 11271              |
-|                            | 16000        | 66.6             | 8954               | 55.3             | 10190              |
-|                            | 12000        | 59.4             | 7927               | 45.6             | 8752               |
-|                            | 8000         | 47.6             | 6282               | 33.8             | 6589               |
 # Inference
 ## vLLM
@@ -59,7 +68,7 @@ sampling_params = SamplingParams(
 )
 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?'

 ---
 base_model:
 - nvidia/OpenMath-Nemotron-14B
+license: cc-by-4.0
+pipeline_tag: text-generation
+library_name: transformers
+tags:
+- mathematical-reasoning
+- qwen2
+paper: https://huggingface.co/papers/2507.08267
 ---
 # Fast-OpenMath-Nemotron-14B
+This model is based on the paper [A Practical Two-Stage Recipe for Mathematical LLMs: Maximizing Accuracy with SFT and Efficiency with Reinforcement Learning](https://huggingface.co/papers/2507.08267).
+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`](https://huggingface.co/RabotniKuma/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`](https://huggingface.co/nvidia/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](https://github.com/analokmaus/kaggle-aimo2-fast-math-r1/tree/master).
+Project page: https://analokmaus.github.io/Fast-Math-R1/
 **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.
 # Evaluation
 <img src='https://github.com/analokmaus/kaggle-aimo2-fast-math-r1/blob/master/assets/pass1_aime_all.png?raw=true' max-height='400px'>
+|                            |              | AIME 2024        |                    | AIME 2025        |                    |
+| -------------------------- | ------------ | ---------------- | ------------------ | ---------------- | ------------------ |
+| Model                      | Token budget | Pass@1 (avg. 64) | Mean output tokens | Pass@1 (avg. 64) | Mean output tokens |
+| OpenMath-Nemotron-14B      | 32000        | 76.2             | 11493              | 64.5             | 13414              |
+|                            | 24000        | 75.4             | 11417              | 63.4             | 13046              |
+|                            | 16000        | 66               | 10399              | 54.2             | 11422              |
+|                            | 12000        | 55               | 9053               | 40               | 9609               |
+|                            | 8000         | 36               | 6978               | 27.2             | 7083               |
+| Fast-OpenMath-Nemotron-14B | 32000        | 70.7             | 9603               | 61.4             | 11424              |
+|                            | 24000        | 70.6             | 9567               | 60.9             | 11271              |
+|                            | 16000        | 66.6             | 8954               | 55.3             | 10190              |
+|                            | 12000        | 59.4             | 7927               | 45.6             | 8752               |
+|                            | 8000         | 47.6             | 6282               | 33.8             | 6589               |
 # Inference
 ## vLLM
 )
 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?'