nvidiaListed Jun 3, 2026

NVIDIA: Nemotron 3 Ultra

NVIDIA Nemotron 3 Ultra is an open frontier-reasoning and orchestration model from NVIDIA, with 55B active parameters out of 550B total (MoE). Built on a hybrid Transformer-Mamba mixture-of-experts architecture, it...

Compare

Context window262.1K

Max output182.5K

Parameters550B

Provider endpoints7

Benchmark results0

Tool calling Structured output Reasoning Prompt caching Open weights

Measured performance

Benchmark results

All leaderboards ↗

—

No published results in this index yet

Scores appear here when an evaluation matches this exact model.

Access & cost

Prices across providers

7 quotes

Your token mix

USD · listed token rates

Input million tokensOutput million tokens

deepinfra · via Hugging Face

$2.25

DeepInfra · via OpenRouterdeepinfra/fp4/flex · tier=flex · quantization=fp4

$0.84

DeepInfra · via OpenRouterdeepinfra/fp4 · quantization=fp4

$1.05

BaseTen · via OpenRouterbaseten/fp4/us · quantization=fp4

$1.20

BaseTen · via OpenRouterbaseten/fp4 · quantization=fp4

$1.20

Venice · via OpenRoutervenice/fp8 · quantization=fp8

$1.4063

OpenRouter

Catalog quote

Input: $0.60
Output: $2.40
Cached input: $0.12

USD / 1M tokens

262.1K context182.5K max outputPrice source ↗Checked Oct 21, 2026

Endpoint ID

nvidia/nemotron-3-ultra-550b-a55b

deepinfra

via Hugging Face

Input: $1.00
Output: $5.00

USD / 1M tokens

262.1K contextPrice source ↗Checked Oct 21, 2026

Endpoint ID

nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16

DeepInfra

via OpenRouter

deepinfra/fp4/flex · tier=flex · quantization=fp4

Input: $0.40
Output: $1.76
Cached input: $0.08

USD / 1M tokens

262.1K context16.4K max outputPrice source ↗Checked Oct 20, 2026

Endpoint ID

nvidia/nemotron-3-ultra-550b-a55b::deepinfra/fp4/flex

DeepInfra

via OpenRouter

deepinfra/fp4 · quantization=fp4

Input: $0.50
Output: $2.20
Cached input: $0.10

USD / 1M tokens

262.1K context16.4K max outputPrice source ↗Checked Oct 20, 2026

Endpoint ID

nvidia/nemotron-3-ultra-550b-a55b::deepinfra/fp4

BaseTen

via OpenRouter

baseten/fp4/us · quantization=fp4

Input: $0.60
Output: $2.40
Cached input: $0.12

USD / 1M tokens

202.8K context182.5K max outputPrice source ↗Checked Oct 20, 2026

Endpoint ID

nvidia/nemotron-3-ultra-550b-a55b::baseten/fp4/us

BaseTen

via OpenRouter

baseten/fp4 · quantization=fp4

Input: $0.60
Output: $2.40
Cached input: $0.12

USD / 1M tokens

202.8K context182.5K max outputPrice source ↗Checked Oct 20, 2026

Endpoint ID

nvidia/nemotron-3-ultra-550b-a55b::baseten/fp4

Venice

via OpenRouter

venice/fp8 · quantization=fp8

Input: $0.625
Output: $3.125
Cached input: $0.1875

USD / 1M tokens

256K context32.8K max outputPrice source ↗Checked Oct 20, 2026

Endpoint ID

nvidia/nemotron-3-ultra-550b-a55b::venice/fp8

Model specifications

Developer: nvidia
Modality: Language
Context: 262,144 tokens
Maximum output: 182,520 tokens
Weights: Open weights
Provider listed: Jun 3, 2026
First discovered: Oct 16, 2026

Exact identities

openrouternvidia/nemotron-3-ultra-550b-a55b
huggingfacenvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16

Sources

OpenRouter ↗Hugging Face ↗

Updated Oct 21, 2026

More from nvidia

View all ↗

nvidiaListed 5 days ago

NVIDIA: Nemotron 3.5 Super VL (free)

NVIDIA Nemotron 3.5 Super VL is an open multimodal mixture-of-experts model from NVIDIA, with 12B active parameters out of 120B total. It takes text, images, documents, GUI screenshots, and video...

1M contextImage inputVideo inputNo reported score

No reported score

Provider degraded →

+Compare

nvidiaListed 5 days ago