Latest Content
Nemotron Cascade 2 30B A3B icon

Nemotron Cascade 2 30B A3B

NVIDIA
Nemotron Cascade 2 30B A3B is an open-weight Mixture-of-Experts (MoE) language model built on Nemotron 3 Nano 30B A3B Base. It contains 30B total parameters with 3B active parameters, activating 6 experts per token across 128 routed experts plus 1 shared expert. The model employs a 52-layer hybrid architecture with Mamba-SSM and MoE layers, 32 attention heads, and a 2,688 hidden size. Supporting a 262K token context window, it introduces multi-domain on-policy distillation within Cascade RL to maximize reasoning, agentic performance, and benchmark recovery across complex tasks.
TypeMoE LLM
CapabilitiesText Generation, Instruction Following, Reasoning, Mathematical Reasoning+5 more
Release Date19 March, 2026
Links
Paper|Blog|HF Model Card
Licensenvidia-open-model-license

Inference Instructions

Deploy and run this model on NVIDIA B200 GPUs using the command below. Copy the command to get started with inference.

CONSOLE
docker run --gpus all 
 --shm-size 128g 
 -p 8000:8000 
 -v ~/.cache/huggingface:/root/.cache/huggingface 
 -e HF_TOKEN='YOUR_HF_TOKEN' 
 --ipc=host 
 lmsysorg/sglang:v0.5.9 
 python3 -m sglang.launch_server 
 --model-path nvidia/Nemotron-Cascade-2-30B-A3B 
 --host 0.0.0.0 
 --port 8000 
 --max-prefill-tokens 65536 
 --mamba-ssm-dtype float32 
 --max-running-requests 1024 
 --tp 8 
 --mem-fraction-static 0.95 
 --tool-call-parser qwen3_coder 
 --reasoning-parser nano_v3 
 --trust-remote-code

Model Benchmarks

Each model was tested with a fixed input size and total token volume while increasing concurrency to measure serving performance under load.

ITL vs Concurrency

Time to First Token

Throughput Scaling

Total Tokens/sec vs Avg TTFT

Vultr Cloud GPU

NVIDIA HGX B200

Deploy NVIDIA B200 on Vultr Cloud GPU

Deploy NVIDIA B200 instancesContact Sales