NVIDIA-Nemotron-Labs-3-Elastic-30B-A3B-BF16

Home/Models/nvidia/NVIDIA-Nemotron-Labs-3-Elastic-30B-A3B-BF16

nvidia/NVIDIA-Nemotron-Labs-3-Elastic-30B-A3B-BF16

kind: original
quant: bf16 (16-bit)
family: nemotron
modality: text-generation
root model: nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16
provider: nvidia
license: other
language: en, es, fr, de, ja, it
context: 128k
library: transformers
vRAM (GB): —
size on disk: —
downloads: 3,604
likes: 30
created: —
last modified: 2026-05-08T03:42:12+00:00 (2026)
huggingface: https://huggingface.co/nvidia/NVIDIA-Nemotron-Labs-3-Elastic-30B-A3B-BF16
tags: transformerssafetensorsnemotron_htext-generationnvidiapytorchelasticconversationalcustom_codeenesfrdejaitarxiv:2512.20848arxiv:2511.16664base_model:nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16base_model:finetune:nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16license:otherendpoints_compatibleregion:us

benchmark	category	score	context	conf	evidence
AIME 2025	math	88.54		medium	`AIME-2025: 88.54`
AIME 2025	math	78.54	AIME-2025	medium	`\| AIME-2025 \| 78.54 \| 85.63 \| 88.54 \| 87.92 \| 80.00 \|`
MMLU-Pro	knowledge	68.28	MMLU-Pro	medium	`\| MMLU-Pro \| 68.28 \| 76.07 \| 78.63 \| 78.86 \| 81.11 \|`
IFBench	instruction-following	67.39	IFBench (instruct)	medium	`\| IFBench (instruct) \| 67.39 \| 70.75 \| 73.96 \| 73.19 \| 46.57 \|`
IFBench	instruction-following	64.03	IFBench (prompt)	medium	`\| IFBench (prompt) \| 64.03 \| 67.43 \| 70.58 \| 70.82 \| 43.28 \|`
GPQA	reasoning	57.39	GPQA	medium	`\| GPQA \| 57.39 \| 69.82 \| 72.10 \| 73.11 \| 70.83 \|`
LiveCodeBench	coding	55.24	LiveCodeBench v5	medium	`\| LiveCodeBench v5 \| 55.24 \| 67.30 \| 72.70 \| 71.75 \| 68.25 \|`

none scraped.