NVIDIA: Nemotron 3 Super

1 provider live

nvidia/nemotron-3-super-120b-a12b

1 live provider
53.5% uptime
1.20 s first token
232 tok/s output

NVIDIA Nemotron 3 Super is a 120B-parameter open hybrid MoE model, activating just 12B parameters for maximum compute efficiency and accuracy in complex multi-agent applications. Built on a hybrid Mamba-Transformer Mixture-of-Experts architecture with multi-token prediction (MTP), it delivers over 50% higher token generation compared to leading open models. The model features a 1M token context window for long-term agent coherence, cross-document reasoning, and multi-step task planning. Latent MoE enables calling 4 experts for the inference cost of only one, improving intelligence and generalization. Multi-environment RL training across 10+ environments delivers leading accuracy on benchmarks including AIME 2025, TerminalBench, and SWE-Bench Verified. Fully open with weights, datasets, and recipes under the NVIDIA Open License, Nemotron 3 Super allows easy customization and secure deployment anywhere — from workstation to cloud.

Context: 256K; tokens
Max output: 16K; tokens
Input: $0.085; per million tokens
Output: $0.40; per million tokens

Providers

Ranked by price and speed. Requests go to the best healthy provider, then fail over down the list.

#	Provider	Input	Output	Cache read	Latency	Throughput	Uptime (3d)
1	DeepInfraNon-EU Data policy not declared	$0.085	$0.40	n/a	1.20 s	232 tok/s	53.5%

Want to serve this model?Become a provider

Price history

No price changes since Oct 20, 2026.

DeepInfra$0.085 in, $0.40 out

Performance

Last 24 hours, from llm.buzz's test requests

Throughput (tok/s)

DeepInfra232 tok/s

Data table

Provider	From (UTC)	p50	p90	p99
DeepInfra	Oct 26, 06:00	232 tok/s	232 tok/s	232 tok/s
DeepInfra	Oct 26, 10:00	232 tok/s	232 tok/s	232 tok/s
DeepInfra	Oct 26, 15:00	193 tok/s	193 tok/s	193 tok/s
DeepInfra	Oct 26, 20:00	93 tok/s	93 tok/s	93 tok/s
DeepInfra	Oct 27, 00:00	112 tok/s	112 tok/s	112 tok/s
DeepInfra	Oct 27, 04:00	278 tok/s	278 tok/s	278 tok/s
DeepInfra	Last 24 hours	232 tok/s	93 tok/s	93 tok/s

Latency (time to first token)

DeepInfra1.20 s

Data table

Provider	From (UTC)	p50	p90	p99
DeepInfra	Oct 26, 06:00	1.44 s	1.44 s	1.44 s
DeepInfra	Oct 26, 10:00	1.20 s	1.20 s	1.20 s
DeepInfra	Oct 26, 15:00	1.20 s	1.20 s	1.20 s
DeepInfra	Oct 26, 20:00	1.44 s	1.44 s	1.44 s
DeepInfra	Oct 27, 00:00	2.48 s	2.48 s	2.48 s
DeepInfra	Oct 27, 04:00	1.20 s	1.20 s	1.20 s
DeepInfra	Last 24 hours	1.20 s	2.48 s	2.48 s

Uptime

Last 72 hours, by hour

Best provider 53.5%

This model

DeepInfra

Up: 95% or more served
Degraded: 85% to 95%
Down: below 85%
No data
From llm.buzz's test requests

Questions

How much does NVIDIA: Nemotron 3 Super cost?

NVIDIA: Nemotron 3 Super costs $0.085 per million input tokens and $0.40 per million output tokens at DeepInfra, the provider requests go to first.

Do providers train on my prompts?

Each provider declares it, and the table above shows every declaration. You can limit a request to providers that keep nothing, or to providers in the EU.

Can I serve NVIDIA: Nemotron 3 Super myself?

Yes. Apply as a provider, add your endpoint and list this model at your price. It goes live once the automated check passes.

Similar models

DeepSeek: DeepSeek V4 Flash 0731deepseek/deepseek-v4-flash-0731

Open weights. 1M context, $0.461 input and $0.9219 output. 3 providers.

OpenAI: gpt-oss-120bopenai/gpt-oss-120b

Open weights. 128K context, $0.05 input and $0.25 output. 3 providers.

Z.ai: GLM 5.3z-ai/glm-5.3

Open weights. 1M context, $0.98 input and $3.08 output. 5 providers.

Call it

import os
from openai import OpenAI

client = OpenAI(
    base_url="https://api.llm.buzz/api/v1",
    api_key=os.environ["LLM_BUZZ_KEY"],
)
reply = client.chat.completions.create(
    model="nvidia/nemotron-3-super-120b-a12b",
    messages=[{"role": "user", "content": "Hello"}],
)

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.llm.buzz/api/v1",
  apiKey: process.env.LLM_BUZZ_KEY,
});
const reply = await client.chat.completions.create({
  model: "nvidia/nemotron-3-super-120b-a12b",
  messages: [{ role: "user", content: "Hello" }],
});

curl https://api.llm.buzz/api/v1/chat/completions \
  -H "Authorization: Bearer $LLM_BUZZ_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "nvidia/nemotron-3-super-120b-a12b",
       "messages": [{"role": "user", "content": "Hello"}]}'

Endpoints

POST /api/v1/chat/completionsPOST /api/v1/responsesPOST /api/v1/messages

Authentication

Send your llm.buzz key as a Bearer token: Authorization: Bearer $LLM_BUZZ_KEY. Create keys in the console. The same key calls every model; each request pays the price of the provider that answered it.

More in Authentication and the API reference.

Call NVIDIA: Nemotron 3 Super with one keyEach request pays the price of the provider that answered it.

Get API key