Gemma-4-26B-A4B-NVFP4

Home/Models/nvidia/Gemma-4-26B-A4B-NVFP4

nvidia/Gemma-4-26B-A4B-NVFP4

kind: nvfp4
quant: nvfp4 (4-bit)
family: gemma
modality: text-generation
root model: google/gemma-4-26B-A4B-it
provider: nvidia
license: apache-2.0
language: —
context: 256K
library: Model Optimizer
vRAM (GB): —
size on disk: —
downloads: 1,185,207
likes: 123
created: —
last modified: 2026-05-11T15:58:52+00:00 (2026)
huggingface: https://huggingface.co/nvidia/Gemma-4-26B-A4B-NVFP4
tags: Model Optimizersafetensorsgemma4nvidiaModelOptquantizedNVFP4nvfp4gemma4-26b-A4B-ittext-generationconversationalbase_model:google/gemma-4-26B-A4B-itbase_model:quantized:google/gemma-4-26B-A4B-itlicense:apache-2.08-bitmodeloptregion:us

benchmark	category	score	context	conf	evidence
IFEval	instruction-following	96.6	IFEval	medium	`\| IFEval \| 96.60% \| 96.40% \|`
AIME 2025	math	88.95	AIME 2025	medium	`\| AIME 2025 \| 88.95% \| 90.00% \|`
MMLU-Pro	knowledge	85.0	MMLU Pro	medium	`\| MMLU Pro \| 85.00% \| 84.80% \|`
LiveCodeBench	coding	80.5	LiveCodeBench (pass@1)	medium	`\| LiveCodeBench (pass@1) \| 80.50% \| 79.80% \|`
GPQA Diamond	reasoning	80.3	GPQA Diamond	medium	`\| GPQA Diamond \| 80.30% \| 79.90% \|`
IFBench	instruction-following	77.77	IFBench	medium	`\| IFBench \| 77.77% \| 78.1% \|`

none scraped.