APIUp to 25% cheaper than official pricesTry the API →
Hermes
APIModelsCases
Back to News
newsLlmsOpen Source

NVIDIA's Nemotron 3.5 Super VL Runs 5x Faster on Agentic Tasks

NVIDIA's new open 120B mixture-of-experts model fuses text, image, document, and video reasoning into one agent-ready backbone with 12B active parameters.

NVIDIA's Nemotron 3.5 Super VL Runs 5x Faster on Agentic Tasks
Source
NVIDIA AI
Published
Author
AlphaSignal Newsroom
Read
1 min read

NVIDIA's new open 120B mixture-of-experts model fuses text, image, document, and video reasoning into one agent-ready backbone with 12B active parameters.

Reporting is indexed from AlphaSignal. Rights remain with the original publisher and cited sources.

Read original report

Next reads

MBZUAI Ships Jais 3, Six Open Arabic Models Beating Qwen and Claude
Institute of Foundation Models · news

MBZUAI Ships Jais 3, Six Open Arabic Models Beating Qwen and Claude

Japan's NII Releases LLM-jp-4 to Beat GPT-4o on Japanese Benchmarks
AlphaSignal · paper

Japan's NII Releases LLM-jp-4 to Beat GPT-4o on Japanese Benchmarks

Huihui-ai Ships Uncensored Qwen3.8 Flash-Next With 176B Parameters
AlphaSignal · model

Huihui-ai Ships Uncensored Qwen3.8 Flash-Next With 176B Parameters