BlogLLM Benchmark Scores as Published, Side by Side

Nemotron 3.5 Super VL 120B-A12B: Benchmark Scores as Published

Every result the catalog holds for Nemotron 3.5 Super VL 120B-A12B: 41 tests in 13 areas, from 1 page. All were printed by NVIDIA itself. Each number is as printed; nothing is ranked or rescored.

Developer NVIDIAReleased 2026-10-15Context window 1MInput Text, image, videoOutput TextWeights Open (OpenMDW License Agreement, version 1.1)Data version 2026103001, as of 2026-10-30

Knowledge Work

Agentic Coding

Science and Math

Long Context

Computer Use

TestValuePrinted bySetup as printed
OSWorld-Verified51.14%NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19-

Multimodal Understanding

General Knowledge and Reasoning

Code Generation

TestValuePrinted bySetup as printed
LiveCodeBench v688.22%NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19-

Tool Use and Workflows

Web Search and Research

Instruction Following

TestValuePrinted bySetup as printed
IFBench70.86NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19-

Multilingual

Writing and Conversation

TestValuePrinted bySetup as printed
LMArena v3 proxy52.67NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19-

More from NVIDIA

Every number here is read from the same catalog file as the benchmark table, data version 2026103001. Where one page printed the model under several setups, each value is listed. The catalog is public in armanobosyan/sugra-llm-benchmarks; a wrong number can be reported from the table, with Report an error under the value, or as an issue in that repository.