models · nvidia

Llama-3.3-Nemotron-Super-49B-v1

NVIDIA pruned + distilled 49B from Llama-3.3-70B via NAS. Near-70B quality at 49B.

What this registry has checked

Quarantine. We hold the publisher’s own published hash for these weights and have recorded it against this entry. We have not downloaded and hashed the weights ourselves, so the hash is the publisher’s claim, not our measurement.

Weight file SHA-256	957f16be72a083739afe6de9d35856ef8a80978d098701b2ad76d30c6d57cdb7
Method	hf-published-sha256

Re-check this yourself → Source → Full record →

What is recorded

Publisher	nvidia
Origin	United States
Licence	Llama-3.3
Publisher’s own licence statement	other
Parameters	49,867,145,216 (49.9B)
Architecture	nemotron-nas
Layers	80
Hidden size	8192
Attention	MHA (64 heads, 64 KV)
Native precision	BF16
Pulls recorded here	42,014
Downloads reported upstream	101,756

Reported scores

MMLU-Pro	69.8
GPQA Diamond	51.7
AIME 2025	7.7
MATH-500	77.5
LiveCodeBench	28
IFBench	39.5
Humanity's Last Exam	3.8

Reported by a third party and recorded here, not re-run by us. Source: artificialanalysis.ai/api/v2.

Elsewhere in the registry

All models → Browse with filters → What will run on my hardware? →

Registry record as of 2026-06-11. Identity, licence and parameter count are properties of a release and do not move; tier can, and the verify link above re-checks it against the live registry rather than this page.