Models/Nemotron-3-Ultra-550b-a55b

Nemotron-3-Ultra-550b-a55bActive · Cloud

Nebius AI Studio · Released Jun 2026 · NVIDIA Open Model License

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...

Context
1.0M
Max output
1.0M
Input
$1.00
Output
$3.00
Cached in
-
Latency
-

Capabilities

Core
Streaming
Function calling
JSON mode
Structured outputs
Multimodal
Vision
Image input
Audio
Image generation
Video
Advanced
Reasoning
Tool calling
MCP
Prompt caching
Batch API
Embeddings

Details

ReleasedJun 2026
LicenseNVIDIA Open Model License
InputText
OutputText
Throughput-
Availability100%

Related models