TPSTokens Per Second

NVIDIA Nemotron 3.5 Super VL 120B-A12B NVFP4

NVIDIA🇺🇸 United States
active
Compare with other models →
Context window1000K tokens

Version History

3.5-super-vl-nvfp4minor

NVIDIA released an NVFP4-quantized checkpoint of Nemotron 3.5 Super VL, a 120B-total/12B-active hybrid Mamba-2 MoE vision-language model, produced via quantization-aware distillation from BF16. NVIDIA reports near-parity with BF16 on most benchmarks, with larger drops on MMLU-ProX, WebVoyager and DeepSWE.

Benchmark Scores

Full leaderboard →
71.1%
SWE-bench Verified

Coverage

model releaseNVIDIA

NVIDIA releases Nemotron 3.5 Super VL NVFP4: 120B-parameter, 12B-active vision-language model with 1M context

NVIDIA has published Nemotron 3.5 Super VL in NVFP4 quantization on Hugging Face: a 120B-parameter, 12B-active hybrid Mamba-2/Transformer MoE with a 1M-token context window and text, image and video input. NVIDIA reports the quantized checkpoint stays close to BF16 on most benchmarks. It fits on one DGX Spark, one H200, two H100s or one GB200.

3 min read
NVIDIANemotronvision-language