Nemotron 3 Ultra
$0.2550% offper 1M tokens$0.550% off
NVIDIA's open model for frontier reasoning and orchestration. It is a hybrid Transformer-Mamba mixture-of-experts with 55B of 550B parameters active.
262KContext
$0.25Input / 1M
$1.10Output / 1M
50%Saved vs retail
client.chat.completions.create( model="nemotron-3-ultra", messages=[ {"role": "user", "content": "Hello"}, ],)
Cost per 1M input tokens
Claw Hunter$0.25
Retail$0.5