Post-Cutoff
Search For AI agents

NVIDIANemotron 3.5 familyCurrentMultimodal modelReleased 15 Oct 2026

NVIDIA Nemotron 3.5 Super VL (120B-A12B)

NVIDIA’s open multimodal model for enterprise document and tool-using agents, released with training recipes and data for customization in NeMo. </content> </invoke>

Model idnvidia/nemotron-3.5-super-vl

Price$0.15 in, $0.50 out per 1M tokens

Checked15 October 2026

No published training cutoff

At least0events since its release on 15 October 2026

0 historic, 0 major. 0 days of news.

NVIDIA doesn’t publish a training cutoff, so we count from its release date. The real gap is larger.

How to call it

4 ways to use NVIDIA Nemotron 3.5 Super VL (120B-A12B), with the model id each provider expects.

Providers, model ids, endpoints and docs for NVIDIA Nemotron 3.5 Super VL (120B-A12B)
ProviderModel idEndpointDocs
Hugging Face huggingface.co
OpenRouter nvidia/nemotron-3.5-super-vl openrouter.ai
NVIDIA NIM / build.nvidia.com (free rate-limited dev API) blogs.nvidia.com
ModelScope blogs.nvidia.com

Pricing

Prices for NVIDIA Nemotron 3.5 Super VL (120B-A12B)
Input$0.15 per 1M tokens
Output$0.50 per 1M tokens

As published: “per 1M tokens (USD) on OpenRouter (also a rate-limited :free variant)”.

Source: openrouter.ai, checked 15 October 2026.

Specs

Input
Text, images and PDF
Output
Text
Context window
1,000,000 tokens
Max output
Not published
Open weights
Yes
Licence
OpenMDW-1.1
Training cutoff
Not published
Released
15 October 2026
Status
Current
Type
Multimodal model

What stands out

What its maker marketed at launch, and what people found later, each with its source.

  • Open vision-language agent model with 1M context

    120B total / 12B active Mamba-2–Transformer hybrid MoE with a C-RADIO vision encoder and native multi-token prediction; reads PDFs, charts, tables, images and web pages; up to 1M tokens.

    Source: huggingface.co

  • Fast at long input

    About 625 output tokens/s at 100K input on four B200s per Artificial Analysis (via VentureBeat); NVIDIA claims up to 5x the throughput of comparable open models.

    Source: venturebeat.com

Notes

Trained on ~26T tokens, 3.8M post-training samples and RL in ~100 open environments. Ranks below GLM-5.3-Flash and Qwen3.8-Flash-Next on the Artificial Analysis index (~26). Production use via NVIDIA AI Enterprise ($4,500/GPU/year).