NVIDIA Nemotron 3 Ultra 550B A55B
NVIDIA Nemotron 3 Ultra 550B A55B is a reasoning model from NVIDIA released on Jun 4, 2026. It reads text and writes text, with a 1M token context window.
Show moreShow less
Its weights are open, so you can host it yourself.
On the Artificial Analysis Intelligence Index it scores 22.9, which places it at number 28 among the 43 models I track with independent scores.
- Modalities
- In / out price
- n/a
- Context
- 1M
- Max output
- n/a
- Released
- Jun 4, 2026
- Knowledge cutoff
- n/a
- Tool calling
- Structured output
- Reasoning
- Prompt caching
- Open weights
- Free to use
Independent benchmarks
Measured by third parties, not reported by NVIDIA. Each links to its source and shows the date I checked it.
A composite of 10 independent evaluations (agents 30%, coding 20%, scientific reasoning 20%, general 30%), run by Artificial Analysis on each model, version 4.3.2.
Artificial Analysis, checked Oct 10, 2026
Human preference: people compare two anonymous answers and pick the better one. Ratings from the Text Arena overall board, style control on.
Arena (formerly LMArena), checked Oct 10, 2026
Median output tokens per second measured by Artificial Analysis on the provider API, for the same model setting as its index score.
Artificial Analysis, checked Oct 10, 2026
Benchmarks NVIDIA reports (8)
- swe-bench-verified (SWE-Bench Verified, Nemo Evaluator SDK)70.7%
- terminal-bench-2.1 (Terminal Bench 2.1, Nemo Evaluator SDK (Harbor))56.4%
- gpqa-diamond (GPQA (no tools))87%
- hle-no-tools (HLE (no tools))26.7%
- hle-with-tools (HLE (with tools))37.4%
- livecodebench (LiveCodeBench (v6))89%
- browsecomp (BrowseComp (with search, internal or official implementation))44.4%
- ruler-1m (RULER (1M))94.7%
Labs pick their own settings and harnesses, so compare these only within one lab.
More from NVIDIA
- 1.05M context
Reads text, images and video. Tool calling, reasoning.
by nvidiaOct 15, 2026$0/M input tokens$0/M output tokens
- Index 12.9
Reads text. Reasoning.
by nvidiaAug 11, 20261M context
- 262K context
Reads text. Tool calling, structured output, reasoning. Served by 11 providers.
by nvidiaAug 11, 2026$0.05/M input tokens$0.20/M output tokens11 providers
- 1M context
Reads text. Tool calling, structured output, reasoning. Served by 20 providers.
by nvidiaJun 4, 2026$0.50/M input tokens$2.50/M output tokens20 providers
- 128K context
Reads text and images. Reasoning. Served by 6 providers.
by nvidiaJun 4, 2026$0.05/M input tokens$0.15/M output tokens6 providers
- 256K context
Reads text, images, audio and video. Tool calling, structured output, reasoning. Served by 7 providers.
by nvidiaApr 28, 2026$0.10/M input tokens$0.25/M output tokens7 providers
NVIDIA Nemotron 3 Ultra 550B A55B questions
What is the context window of NVIDIA Nemotron 3 Ultra 550B A55B?
1,000,000 tokens of input. As a rough guide, a million tokens is around 750,000 English words.
Does NVIDIA Nemotron 3 Ultra 550B A55B support tool calling?
Not through the standard tools parameter. For an agent that calls your APIs, pick a model from the alternatives below that does.
Is NVIDIA Nemotron 3 Ultra 550B A55B open source?
Its weights are published, so you can run it on your own hardware or a cloud of your choice. Read the license before commercial use.
When was NVIDIA Nemotron 3 Ultra 550B A55B released?
Jun 4, 2026.
How good is NVIDIA Nemotron 3 Ultra 550B A55B compared with other models?
It scores 22.9 on the Artificial Analysis Intelligence Index and 1426 on the Arena text leaderboard. Benchmarks are a shortlist, not a verdict: test the top two or three candidates on your own tasks before you commit.
Catalog data from models.dev (MIT licensed), refreshed daily, last on Oct 15, 2026. Independent scores from my hand checked rankings, sources linked above. The summary is written from these facts.
Building on NVIDIA Nemotron 3 Ultra 550B A55B?
I test your real tasks on it and the two closest alternatives, then ship the agent on whichever wins on cost and accuracy.
Book Meeting