Models / NVIDIA

NVIDIA

Deploy the latest NVIDIA Nemotron models on Together AI. Open reasoning models built for multi-agent orchestration, with long context and a hybrid Mamba-Transformer architecture.

Why NVIDIA on Together AI?

Designed for production workloads that need 
consistent performance and operational control.

Engineered for agents at throughput

Nemotron's hybrid Transformer and Mamba architecture keeps active parameters low while maintaining strong reasoning, and multi-token prediction generates several tokens per forward pass to speed up long outputs.

Fully open, with cost control built in

Nemotron is fully open under the NVIDIA Open Model License, with open weights, training data, and recipes. Token budget supports cost-controlled reasoning for predictable accuracy and spend.

Enterprise-ready from day one

SOC 2 Type II certified, HIPAA compliant, and deployed on US-based infrastructure. Full model ownership with no data retention by default.

Meet the NVIDIA family

Explore top-performing models across text, image, video, code, and voice.

Chat

NVIDIA Nemotron 3.5 Lightning

Chat

NVIDIA Nemotron 3 Ultra

Transcribe

NVIDIA Parakeet TDT 0.6B V3 Realtime

Transcribe

NVIDIA Nemotron 3 ASR Streaming 0.6B

Transcribe

NVIDIA Parakeet TDT 0.6B v3

Chat

NVIDIA Nemotron 3 Super

Chat

NVIDIA Nemotron 3 Nano Omni

Transcribe

NVIDIA Nemotron 3.5 ASR

Chat

NVIDIA Nemotron 3 Nano

Chat

NVIDIA-Nemotron-Nano-9B-v2

Deployment options

Run models using different deployment options depending on latency needs, traffic patterns, and infrastructure control.

  • Serverless Inference

  • Provisioned 
Throughput

  • Dedicated Model 
Inference

  • Dedicated Container 
Inference

Serverless Inference

A fully managed real-time or batch inference API with access to dozens of the most popular AI models.

Best for

Variable or unpredictable traffic

Rapid prototyping and iteration

Cost-sensitive or early-stage production workloads

Provisioned 
Throughput

Reserved token capacity with SLA guarantees. Priced in PTUs, a normalized throughput unit. 

Best for

Production workloads

Reliability guarantees

Predictable pricing

Dedicated Model 
Inference

An inference endpoint backed by reserved, isolated compute resources and Together AI inference research.

Best for

Predictable or steady traffic

Latency-sensitive applications

High-throughput production workloads

Dedicated Container 
Inference

Run inference with your own engine and model on fully-managed, scalable infrastructure.

Best for

Generative media models

Non-standard runtimes

Custom inference pipelines