Nemotron-Cascade-8B-Thinking

Home/Models/nvidia/Nemotron-Cascade-8B-Thinking

nvidia/Nemotron-Cascade-8B-Thinking

kind: original
quant: bf16 (16-bit)
family: nemotron
modality: text-generation
root model: nvidia/Nemotron-Cascade-8B-Thinking
provider: nvidia
license: other
language: en
context: 90K
library: transformers
vRAM (GB): —
size on disk: —
downloads: 440
likes: 41
created: —
last modified: 2026-01-01T01:26:44+00:00 (2026)
huggingface: https://huggingface.co/nvidia/Nemotron-Cascade-8B-Thinking
tags: transformerssafetensorsqwen3text-generationnvidiaNemotron-Cascadereasoninggeneral-purposeSFTRLpytorchconversationalenarxiv:2512.13607arxiv:2309.00071license:othertext-generation-inferenceendpoints_compatibleregion:us

benchmark	category	score	context	conf	evidence
Chatbot Arena	chat	85.8	ArenaHard	medium	`\| ArenaHard \| 85.8 \| 74.6 \| 95.1 \| 95.7 \| 85.8 \| 87.9 \|`
IFEval	instruction-following	85.0	IFEval (Strict Prompt)	medium	`\| IFEval (Strict Prompt) \| 85.0 \| 86.1 \| 84.1 \| 89.8 \| 83.7 \| 90.2 \|`
MMLU	knowledge	83.0	MMLU	medium	`\| MMLU \| 83.0 \| 82.6 \| 89.9 \| - \| 84.0 \| 83.7 \|`
AIME 2024	math	76.0	AIME 2024	medium	`\| AIME 2024 \| 76.0 \| 81.9 \| 91.4 \| 82.3 \| 88.8 \| 89.5 \|`
MMLU-Pro	knowledge	75.1	MMLU Pro	medium	`\| MMLU Pro \| 75.1 \| 73.3 \| 85.0 \| 81.9 \| 75.5 \| 75.7 \|`
BFCL (function calling)	agentic	68.1	BFCL V3	medium	`\| BFCL V3 \| 68.1 \| 66.9 \| 67.9 \| 68.6 \| 67.0 \| 64.4 \|`
AIME 2025	math	67.3	AIME 2025	medium	`\| AIME 2025 \| 67.3 \| 72.0 \| 87.5 \| 72.0 \| 81.4 \| 80.1 \|`
GPQA Diamond	reasoning	62.0	GPQA-Diamond	medium	`\| GPQA-Diamond \| 62.0 \| 64.0 \| 81.0 \| 82.8 \| 66.7 \| 66.5 \|`
IFBench	instruction-following	34.4	IFBench	medium	`\| IFBench \| 34.4 \| 37.4 \| 38.0 \| 36.1 \| 41.4 \| 40.8 \|`

none scraped.