nvidia/nemotron-3-ultra-550b-a55bNVIDIA
nvidia/nemotron-3-ultra-550b-a55b
Input150 DA
Output660 DA
Context350K
Nemotron 3 Ultra NVFP4
ReasoningTool use
NVIDIA's frontier-scale hybrid Latent-MoE model (550B total, 55B active) for the most demanding agentic, reasoning, and long-context workloads across code, math, and science. Trained with an NVFP4 recipe.
Model id — paste this in your code
nvidia/nemotron-3-ultra-nvfp4Pricing · per 1M tokens
Show prices in
Input150 DA$0.6
Output660 DA$2.64
Cached input30 DA$0.12
Context window262KMax output: 66K
Capabilities
- Reasoning — Thinks step by step before answering
- Tool use — Can call your functions / tools
Estimate your cost
Estimated cost
Use this model
Create an account, top up in dinars and call it with your key:
from openai import OpenAI
client = OpenAI(
base_url="https://dzrouter.com/v1",
api_key="sk-dz-…",
)
reply = client.chat.completions.create(
model="nvidia/nemotron-3-ultra-nvfp4",
messages=[{"role": "user", "content": "Salam!"}],
)
print(reply.choices[0].message.content)import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://dzrouter.com/v1",
apiKey: process.env.DZROUTER_API_KEY,
});
const reply = await client.chat.completions.create({
model: "nvidia/nemotron-3-ultra-nvfp4",
messages: [{ role: "user", content: "Salam!" }],
});
console.log(reply.choices[0].message.content);curl https://dzrouter.com/v1/chat/completions \
-H "Authorization: Bearer $DZROUTER_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "nvidia/nemotron-3-ultra-nvfp4",
"messages": [{"role": "user", "content": "Salam!"}],
"stream": true
}'from anthropic import Anthropic
client = Anthropic(
base_url="https://dzrouter.com",
api_key="sk-dz-…",
)
msg = client.messages.create(
model="nvidia/nemotron-3-ultra-nvfp4",
max_tokens=1024,
messages=[{"role": "user", "content": "Salam!"}],
)
print(msg.content[0].text)More from NVIDIA
nvidia/nemotron-3-nano-30b-a3bNVIDIA
nvidia/nemotron-3-nano-30b-a3b
Input15 DA
Output60 DA
Context350K
nvidia/nemotron-3-super-120b-a12bNVIDIA
nvidia/nemotron-3-super-120b-a12b
Input25.5 DA
Output120 DA
Context350K
Nemotron 3 120B A12BNVIDIA
nvidia/nemotron-3-120b
NVIDIA's hybrid MoE model with leading accuracy for multi-agent applications and specialized agentic AI systems.
Input25.5 DA
Output120 DA
Context262K
Nemotron 3 Nano 30B A3BNVIDIA
nvidia/nemotron-3-nano-30b
NVIDIA's efficient hybrid MoE model combining Mamba-2 and Transformer layers, with 30B total and 3.5B active parameters. Reasoning can be toggled on or off, and it handles English, code and several European and Asian languages.
Input15 DA
Output60 DA
Context262K
Nemotron 3 Nano Omni 30B A3B ReasoningNVIDIA
nvidia/nemotron-3-nano-omni
NVIDIA's compact multimodal Nemotron 3 Nano with 30B total / 3B active MoE parameters and built-in reasoning. Strong at agentic and analytical tasks.
Input90 DA
Output549 DA
Context262K
Start building with AI today
Create your account, top up from 500 DA and send your first request in minutes.
Create a free account