OpenReasoning-Nemotron-32B

Home/Models/nvidia/OpenReasoning-Nemotron-32B

nvidia/OpenReasoning-Nemotron-32B

kind: original
quant: bf16 (16-bit)
family: nemotron
modality: text-generation
root model: Qwen/Qwen2.5-32B
provider: nvidia
license: cc-by-4.0
language: en
context: —
library: transformers
vRAM (GB): —
size on disk: —
downloads: 9,429
likes: 125
created: —
last modified: 2025-09-16T16:59:34+00:00 (2025)
huggingface: https://huggingface.co/nvidia/OpenReasoning-Nemotron-32B
tags: transformerssafetensorsqwen2text-generationnvidiacodeconversationalenarxiv:2504.16891arxiv:2507.09075arxiv:2504.01943base_model:Qwen/Qwen2.5-32Bbase_model:finetune:Qwen/Qwen2.5-32Blicense:cc-by-4.0text-generation-inferenceendpoints_compatibleregion:us

benchmark	category	score	context	conf	evidence
SciCode	coding	100.0		low	`\| 1.5B \| 31.0 \| 31.6 \| 47.5 \| 5.5 \| 28.6 \| 1.0 \| 55.5 \| 45.6 \| 31.5 \|`
AIME 2024	math	55.5		low	`\| 1.5B \| 31.0 \| 31.6 \| 47.5 \| 5.5 \| 28.6 \| 1.0 \| 55.5 \| 45.6 \| 31.5 \|`
AIME 2024	math	55.5	AIME24	medium	`\| AIME24 \| 55.5 \| 76.7 \| 76.7 \|`
MMLU-Pro	knowledge	47.5		low	`\| 1.5B \| 31.0 \| 31.6 \| 47.5 \| 5.5 \| 28.6 \| 1.0 \| 55.5 \| 45.6 \| 31.5 \|`
AIME 2025	math	45.6		low	`\| 1.5B \| 31.0 \| 31.6 \| 47.5 \| 5.5 \| 28.6 \| 1.0 \| 55.5 \| 45.6 \| 31.5 \|`
AIME 2025	math	45.6	AIME25	medium	`\| AIME25 \| 45.6 \| 70.0 \| 70.0 \|`
GPQA	reasoning	31.6		low	`\| 1.5B \| 31.0 \| 31.6 \| 47.5 \| 5.5 \| 28.6 \| 1.0 \| 55.5 \| 45.6 \| 31.5 \|`
LiveCodeBench	coding	28.6		low	`\| 1.5B \| 31.0 \| 31.6 \| 47.5 \| 5.5 \| 28.6 \| 1.0 \| 55.5 \| 45.6 \| 31.5 \|`
Humanity's Last Exam	knowledge	11.8	HLE	medium	`\| HLE \| 11.8 \| 13.4 \| 15.5 \|`
Humanity's Last Exam	knowledge	5.5		low	`\| 1.5B \| 31.0 \| 31.6 \| 47.5 \| 5.5 \| 28.6 \| 1.0 \| 55.5 \| 45.6 \| 31.5 \|`

none scraped.