All Models

NVIDIA AI Models

Builds Nemotron models optimized for NVIDIA hardware. Leading GPU maker powering most AI training.

Founded 1993Santa Clara, CA7 ModelsWebsite →
Nvidia logo

Nemotron 3 Ultra

Nvidia

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...

Context512K
Speed215 tok/s
InputText
OutputText
ReasoningYes
NVIDIA logo

Nemotron 3 Ultra 550B A55B (Reasoning)

NVIDIA

A powerful AI model for general-purpose tasks.

ContextN/A
Speed215 tok/s
InputN/A
OutputN/A
ReasoningNo
NVIDIA logo

Nemotron 3 Nano Omni 30B A3B Reasoning

NVIDIA

A powerful AI model for general-purpose tasks.

ContextN/A
Speed319 tok/s
InputN/A
OutputN/A
ReasoningNo
NVIDIA logo

Nemotron Cascade 2 30B A3B

NVIDIA

A powerful AI model for general-purpose tasks.

ContextN/A
SpeedN/A
InputN/A
OutputN/A
ReasoningNo
NVIDIA logo

Nemotron 3 Nano 4B

NVIDIA

A powerful AI model for general-purpose tasks.

ContextN/A
SpeedN/A
InputN/A
OutputN/A
ReasoningNo
Nvidia logo

Nemotron 3 Super

Nvidia

NVIDIA Nemotron 3 Super is a 120B-parameter open hybrid MoE model, activating just 12B parameters for maximum compute efficiency and accuracy in complex multi-agent applications. Built on a hybrid Mamba-Transformer...

Context262K
Speed172 tok/s
InputText
OutputText
ReasoningYes
NVIDIA logo

Nemotron 3 Super 120B A12B (Reasoning)

NVIDIA

A powerful AI model for general-purpose tasks.

ContextN/A
Speed172 tok/s
InputN/A
OutputN/A
ReasoningNo