Nemotron-3-Nano-30B-A3B

Home/Models/unsloth/Nemotron-3-Nano-30B-A3B

Nemotron-3-Nano-30B-A3B

unsloth/Nemotron-3-Nano-30B-A3B

kind: original
quant: bf16 (16-bit)
family: nemotron
modality: text-generation
root model: nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16
provider: unsloth
license: other
language: en, es, fr, de, ja, it
context: 256k
library: transformers
vRAM (GB): —
size on disk: —
downloads: 1,604
likes: 14
created: —
last modified: 2026-03-26T11:45:08+00:00 (2026)
huggingface: https://huggingface.co/unsloth/Nemotron-3-Nano-30B-A3B
tags: transformerssafetensorsnvidiaunslothtext-generationconversationalenesfrdejaitdataset:nvidia/Nemotron-Pretraining-Code-v1dataset:nvidia/Nemotron-CC-v2dataset:nvidia/Nemotron-Pretraining-SFT-v1dataset:nvidia/Nemotron-CC-Math-v1dataset:nvidia/Nemotron-Pretraining-Code-v2dataset:nvidia/Nemotron-Pretraining-Specialized-v1dataset:nvidia/Nemotron-CC-v2.1dataset:nvidia/Nemotron-CC-Code-v1dataset:nvidia/Nemotron-Pretraining-Dataset-sampledataset:nvidia/Nemotron-Competitive-Programming-v1dataset:nvidia/Nemotron-Math-v2dataset:nvidia/Nemotron-Agentic-v1

Benchmark scores (25)

benchmark	category	score	context	conf	evidence
AIME 2025	math	99.2	AIME25 (with tools)	medium	`\| AIME25 (with tools) \| 99.2 \| \- \| 98.7 \|`
RULER	long-context	92.9	RULER-100@256k	medium	`\| RULER-100@256k \| 92.9 \| 89.4 \| \- \|`
RULER	long-context	91.3	RULER-100@512k	medium	`\| RULER-100@512k \| 91.3 \| 84.0 \| \- \|`
AIME 2025	math	89.1	AIME25 (no tools)	medium	`\| AIME25 (no tools) \| 89.1 \| 85.0 \| 91.7 \|`
RULER	long-context	86.3	RULER-100@1M	medium	`\| RULER-100@1M \| 86.3 \| 77.5 \| \- \|`
miniF2F	math	79.9	MiniF2F pass@32	medium	`\| MiniF2F pass@32 \| 79.9 \| 16.8 \| 43.0 \|`
MMLU-Pro	knowledge	78.3	MMLU-Pro	medium	`\| MMLU-Pro \| 78.3 \| 80.9 \| 75.0 \|`
GPQA	reasoning	75.0	GPQA (with tools)	medium	`\| GPQA (with tools) \| 75.0 \| \- \| 74.2 \|`
GPQA	reasoning	73.0	GPQA (no tools)	medium	`\| GPQA (no tools) \| 73.0 \| 73.4 \| 71.5 \|`
Chatbot Arena	chat	72.1	Arena-Hard-V2 (Hard Prompt)	medium	`\| Arena-Hard-V2 (Hard Prompt) \| 72.1 \| 49.6\* \| 71.2\* \|`
IFBench	instruction-following	71.5	IFBench (prompt)	medium	`\| IFBench (prompt) \| 71.5 \| 51.0 \| 65.0 \|`
LiveCodeBench	coding	68.3	LiveCodeBench (v6 2025-08–2025-05)	medium	`\| LiveCodeBench (v6 2025-08–2025-05) \| 68.3 \| 66.0 \| 61.0 \|`
Chatbot Arena	chat	67.7	Arena-Hard-V2 (Average)	medium	`\| Arena-Hard-V2 (Average) \| 67.7 \| 57.8 \| 48.6 \|`
Chatbot Arena	chat	63.2	Arena-Hard-V2 (Creative Writing)	medium	`\| Arena-Hard-V2 (Creative Writing) \| 63.2 \| 66.0\* \| 25.9& \|`
MMLU-Pro	knowledge	59.5	MMLU-ProX (avg over langs)	medium	`\| MMLU-ProX (avg over langs) \| 59.5 \| 77.6\* \| 69.1\* \|`
tau-bench	agentic	56.9	TauBench V2 (Retail)	medium	`\| TauBench V2 (Retail) \| 56.9 \| 58.8 \| 38.0 \|`
BFCL (function calling)	agentic	53.8	BFCL v4	medium	`\| BFCL v4 \| 53.8 \| 46.4\* \| \- \|`
miniF2F	math	50.0	MiniF2F pass@1	medium	`\| MiniF2F pass@1 \| 50.0 \| 5.7 \| 12.1 \|`
tau-bench	agentic	49.0	TauBench V2 (Average)	medium	`\| TauBench V2 (Average) \| 49.0 \| 47.7 \| 48.7 \|`
tau-bench	agentic	48.0	TauBench V2 (Airline)	medium	`\| TauBench V2 (Airline) \| 48.0 \| 58.0 \| 38.0 \|`
tau-bench	agentic	42.2	TauBench V2 (Telecom)	medium	`\| TauBench V2 (Telecom) \| 42.2 \| 26.3 \| 49.7 \|`
SciCode	coding	33.3	SciCode (subtask)	medium	`\| SciCode (subtask) \| 33.3 \| 33.0 \| 34.0 \|`
Humanity's Last Exam	knowledge	15.5	HLE (with tools)	medium	`\| HLE (with tools) \| 15.5 \| \- \| 17.3 \|`
Humanity's Last Exam	knowledge	10.6	HLE (no tools)	medium	`\| HLE (no tools) \| 10.6 \| 9.8 \| 10.9 \|`
Terminal-Bench	agentic-coding	8.5	Terminal Bench (hard subset)	medium	`\| Terminal Bench (hard subset) \| 8.5 \| 5.0 \| 6.0 \|`

Discussions (0)

none scraped.