Nemotron 3.5 Super VL

Name: Nemotron 3.5 Super VL model facts
Creator: OSMEXO

nvidia/nemotron-3.5-super-vl

NVIDIA
Family · Nemotron 3 Super VL
Released 15 Oct 2026
Text · Image · Video → Text
Tools · JSON · Vision

Not on OSMEXO yet

Request this model

Spec & list price

Context1M

Max output236K

Input / 1M$0.15

Output / 1M$0.503.3× input

Scores · standing

Intelligence26.5#83/489

LiveBench50.2#28/32

Usage

Tokens / day419.0B▲ 46% 7d

Usage rank#13

Requests / day735.8K

Tool calls / req0.32

Per request · 30 days

Avg input23.3K1.6× median

Avg output1.2K1.8× median

Cache hit77%about median

Reasoning share82%1.4× median

Tool-call failures3.1%1.7× median

Consumption

Peak 47.2B 29 OctLow 3.4B 15 OctRank today #13

News & trends

17 mentions this week ▲ 89%this model · videos, posts, articles · 30 days

Trending · 14 days

@arcprize on X14.8K views
NVIDIA's New 120B AI Model Is INSANE! 🤯 Nemotron 3.5 Super VL2.7K views
NEW Nemotron 3.5 Super VL + Hermes Agent is Crazy Good! 😱4.2K views
@AI21Labs on X800 views

News

Accelerate agentic AI with NVIDIA Nemotron models on Amazon EKS and SageMaker HyperPod EKSAWS · 29 Oct
NVIDIA Open Models Advance Europe’s Sovereign Intelligence Across IndustriesNVIDIA · 21 Oct
NVIDIA Nemotron 3.5 Super VL Powers Specialized Agents with Multimodal ReasoningNVIDIA · 15 Oct
Open From Data Center to Desktop: NVIDIA Expands Nemotron Family With Highest Efficiency Super VL and Bolt ModelsNVIDIA · 15 Oct

Benchmarks

bar = share of the top score · Benchmark data: Artificial Analysis

Long-context reasoning#153 of 393

69.0%

SciCode#71 of 160

47.8%

Humanity's Last Exam#90 of 465

30.7%

Intelligence index#83 of 489

26.5

Terminal-Bench 4.0#86 of 152

0.0%

Speed · median across 10 providers, latest day

Output speed#19/334181 tok/s2.7× faster than median

Fastest provider470 tok/sbest of 10 providers

First token#590.51 s2.0× quicker than median

First answer token#364.92 sthinking included

#4 cheapest of 26 with a similar score

Value

Intelligence Index

Blended price per 1M tokens (log)

LiveBench suites

Browser21.6

Coding71.3

Data Analysis64.6

Design62.8

Long Agentic11.7

Reasoning74.6

Research66.7

▲ Strongest: Reasoning, Coding. ▼ Weakest: Long Agentic (12).

Technical details

Tokenizer: Other
Moderation: none

Request parameters 19

tools
tool_choice
response_format
structured_outputs
reasoning
include_reasoning
temperature
top_p
top_k
min_p

+9 more

max_tokens
stop
seed
frequency_penalty
logit_bias
logprobs
presence_penalty
repetition_penalty
top_logprobs

Interesting facts

1.7× the output room of similar models (235.9K against a 139.3K median).
Released 18 days ago.

When to use

Best for

Image and video reasoning agents
Document and chart understanding
Research-style long context analysis
Multimodal tool-calling assistants

Think twice for

Terminal coding agents
Browser automation agents
Self-hosting (closed weights)

Research papers

Nemotron 3.5 Bolt 74B-A9B 24GB: A Hybrid Vision-Language Model for Local AI15 Oct

Alternatives

other vendors · similar capability

Tencent HY3
Tencent
25.3
MiniMax M3
MiniMax
29.2
Solar Pro 4
Upstage
26.6
DeepSeek V3.2
DeepSeek
21.5

Want Nemotron 3.5 Super VL on OSMEXO?

Request this model on OSMEXO