VENICE ATLAS · MODEL DOSSIER · NVIDIA NEMOTRON 3.5 SUPER VL 120B

Model overview

NVIDIA Nemotron 3.5 Super VL 120B

Name: NVIDIA Nemotron 3.5 Super VL 120B
Brand: venice.ai
Price: 0.2500 USD
Availability: InStock

NVIDIA Nemotron 3.5 Super VL is an open multimodal vision-language model for specialized agents. Its 120B-parameter Mixture-of-Experts design (12B active) understands text, images, documents, GUIs, and video, combining reasoning, instruction following, and tool use for long-running agentic workflows.

nvidia-nemotron-3-5-super-vl-120b-a12b · venice.ai

Text · Venice catalog$0.25 per 1M tokensPrice on Venice AI

ModalityTextOn Venice AI

Primary price$0.25 per 1M tokensper 1M tokens

Context window262KMax tokens

Benchmarks19Benchmark results

Benchmark results

See the tests behind each score. They use the benchmark’s original scoring scale.

Biology#2532.13 rating Chemistry#2732.28 rating Image To Text#522.73 rating Knowledge#910.92 rating Long Context#1322.91 rating Physics#2832.28 rating Science#1010.92 rating Search#2318.54 rating

Agents1 result

Tau3 Bankingtau3-banking · agents

18.9%

ReportedOct 2026

General intelligence1 result

GDPval-AAgdpval-aa · general

40.4%

ReportedOct 2026

Instruction Following1 result

IFBenchifbench · instruction_following

70.9%

ReportedOct 2026

Math1 result

MathVisionmathvision · math

78.2%

ReportedOct 2026

Multimodal1 result

OSWorld-Verifiedosworld-verified · multimodal

51.1%

ReportedOct 2026

Reasoning11 results

AA-LCRaa-lcr · reasoning

68.6%

ReportedOct 2026

BrowseCompbrowsecomp · reasoning

78.0%

ReportedOct 2026

CharXiv-Rcharxiv-r · reasoning

80.8%

ReportedOct 2026

GPQAgpqa · reasoning

89.5%

ReportedOct 2026

Humanity's Last Examhumanity's-last-exam · reasoning

31.8%

ReportedOct 2026

LiveCodeBench v6livecodebench-v6 · reasoning

88.2%

ReportedOct 2026

SciCodescicode · reasoning

47.1%

ReportedOct 2026

SWE-bench Multilingualswe-bench-multilingual · reasoning

67.7%

ReportedOct 2026

SWE-Bench Proswe-bench-pro · reasoning

51.8%

ReportedOct 2026

SWE-Bench Verifiedswe-bench-verified · reasoning

71.1%

ReportedOct 2026

Terminal-Bench 2.1terminal-bench-2.1 · reasoning

66.7%

ReportedOct 2026

Science1 result

OmniScienceomniscience · science

26.2%

ReportedOct 2026

Vision2 results

MMLongBench-Docmmlongbench-doc · vision

62.8%

ReportedOct 2026

OCRBench-V2 (en)ocrbench-v2-(en) · vision

68.1%

ReportedOct 2026

Learn what each benchmark measures ↗

Pricing detail

Input price per 1M tokens: $0.25
Output price per 1M tokens: $1.25
Cache read price per 1M tokens: $0.13
Blended price per 1M tokens (50/50): $0.75

Prices are published by Venice AI. Venice model page ↗

Capabilities

ReasoningFunction callingVisionVideo inputWeb search

Endpoint metadata

Privacy: private
Quantization: fp4
Maximum completion tokens: 33K
Uncensored: No
Beta: No
Mapping confidence: 100.0%

View model source ↗

All capabilities

Max Images: 10
Max Videos: 1
Quantization: fp4
Supports E2 E E: No
Supports Vision: Yes
Supports X Search: No
Optimized For Code: Yes
Supports Log Probs: No
Supports Reasoning: Yes
Supports Web Search: Yes
Supports Audio Input: No
Supports Video Input: Yes
Default Reasoning Effort: low
Reasoning Effort Options: none, low, medium, high
Supports Multiple Images: Yes
Supports Response Schema: Yes
Supports Tee Attestation: No
Supports Function Calling: Yes
Supports Reasoning Effort: Yes

Related models

More from nvidia

nvidia-nemotron-3-ultra-550b-a55bNVIDIA Nemotron 3 Ultra$0.63 input per 1M tokens ↗nvidia-nemotron-3-nano-30b-a3bNVIDIA Nemotron 3 Nano 30B$0.07 input per 1M tokens ↗