AceMath-RL-Nemotron-7B

Home/Models/nvidia/AceMath-RL-Nemotron-7B

nvidia/AceMath-RL-Nemotron-7B

kind: original
quant: bf16 (16-bit)
family: nemotron
modality: text-generation
root model: nvidia/AceMath-RL-Nemotron-7B
provider: nvidia
license: other
language: en
context: —
library: transformers
vRAM (GB): —
size on disk: —
downloads: 102
likes: 26
created: —
last modified: 2025-04-23T18:26:30+00:00 (2025)
huggingface: https://huggingface.co/nvidia/AceMath-RL-Nemotron-7B
tags: transformerssafetensorsqwen2text-generationnvidiareasoningmathreinforcement learningpytorchconversationalenlicense:othertext-generation-inferenceendpoints_compatibleregion:us

benchmark	category	score	context	conf	evidence
MATH-500	math	92.8		low	`\| DeepSeek-R1-Distill-Qwen-7B \| 92.7 \| 92.8 \| 57.4 \| 82.3 \| 58.2 \| 56.7 \| 89.0 \| 37.6 \|`
GSM8K	math	92.7		low	`\| DeepSeek-R1-Distill-Qwen-7B \| 92.7 \| 92.8 \| 57.4 \| 82.3 \| 58.2 \| 56.7 \| 89.0 \| 37.6 \|`
GAOKAO	knowledge	82.3		low	`\| DeepSeek-R1-Distill-Qwen-7B \| 92.7 \| 92.8 \| 57.4 \| 82.3 \| 58.2 \| 56.7 \| 89.0 \| 37.6 \|`
MATH	math	69.8	Skywork-OR1-Math-7B-Preview	medium	`\| Skywork-OR1-Math-7B-Preview \| 69.8 \| 52.3 \| - \|`
Math Olympiad	math	58.2		low	`\| DeepSeek-R1-Distill-Qwen-7B \| 92.7 \| 92.8 \| 57.4 \| 82.3 \| 58.2 \| 56.7 \| 89.0 \| 37.6 \|`
MATH	math	57.4		low	`\| DeepSeek-R1-Distill-Qwen-7B \| 92.7 \| 92.8 \| 57.4 \| 82.3 \| 58.2 \| 56.7 \| 89.0 \| 37.6 \|`
AIME 2024	math	55.5		low	`\| DeepSeek-R1-Distill-Qwen-7B \| 55.5 \| 39.2 \| 49.1 \|`
GPQA Diamond	reasoning	49.1		low	`\| DeepSeek-R1-Distill-Qwen-7B \| 55.5 \| 39.2 \| 49.1 \|`
AIME 2025	math	39.2		low	`\| DeepSeek-R1-Distill-Qwen-7B \| 55.5 \| 39.2 \| 49.1 \|`
LiveCodeBench	coding	37.6		low	`\| DeepSeek-R1-Distill-Qwen-7B \| 92.7 \| 92.8 \| 57.4 \| 82.3 \| 58.2 \| 56.7 \| 89.0 \| 37.6 \|`

none scraped.