Nemotron 3 Ultra

NVIDIA · 560.5B parameters · Language models

Weights and memory

1 option
FormatFile sizeYour deviceLink
1121 GB1045 GB ◇Hugging Face

Published by the lab. File size is what the repository lists; for original weights it is the total of the weight files, which can include more than one variant. ● memory stated by the publisher · ◇ estimate from file size, an 8K-token context and 0.5 GB for the runtime. Unknown means there is no figure to go on. Checked 2026-10-11.

Tests on this site

Not tested yet

This site has not run this model. What will be tested

Third-party scores

Measured by other sites on hosted endpoints. A quantized file you run yourself may score differently.

OpenRouter evals

data from 2026-10-10
EndpointBenchmarkScoreTasks run
nvidia/nemotron-3-ultra-550b-a55b-2026060481.2% ± 0.0172
nvidia/nemotron-3-ultra-550b-a55b-2026060472.0% ± 0.050

Source: OpenRouter evals (openrouter.ai) via OpenRouter (openrouter.ai/rankings). openrouter.ai

Artificial Analysis

data from 2026-10-09
EndpointIntelligenceCodingAgentic
Nemotron 3 Ultra 550B A55B (Reasoning)22.949.320.1

Source: Artificial Analysis (artificialanalysis.ai) via OpenRouter (openrouter.ai/rankings). artificialanalysis.ai

Design Arena

data from 2026-10-11
EndpointCategoryWin rate
nvidia/nemotron-3-ultra-550b-a55b-20260604codecategories114936.1%
nvidia/nemotron-3-ultra-550b-a55b-20260604uicomponent114536.4%
nvidia/nemotron-3-ultra-550b-a55b-20260604website114534.9%
nvidia/nemotron-3-ultra-550b-a55b-20260604dataviz114437.4%
nvidia/nemotron-3-ultra-550b-a55b-20260604gamedev114436.4%
nvidia/nemotron-3-ultra-550b-a55b-202606043d113539.2%
nvidia/nemotron-3-ultra-550b-a55b-20260604asciiart110438.1%
nvidia/nemotron-3-ultra-550b-a55b-20260604svg106832.5%

Source: Design Arena (www.designarena.ai) via OpenRouter (openrouter.ai/rankings). designarena.ai