Speed
Output tokens / second
Nemotron 3 Super 120B-A12B overview
Nemotron 3 Super 120B-A12B by NVIDIA
A hybrid Mixture-of-Experts reasoning model that punches far above its active parameter count, running on just 12 billion active weights while drawing on the depth of 120 billion total parameters. Built for the realities of production-grade agentic systems.
Nemotron 3 Super 120B-A12B specs
- Model ID
nvidia/nemotron-3-super-120b-a12b- Developer
- NVIDIA
- Context window
- 262,144 tokens
- Max output
- 262,144 tokens
- Input
- Text, Image
- Output
- Text
- API
- OpenAI-compatible, base URL
https://api.aimlapi.com/v1
What Is Nemotron 3 Super?
NVIDIA's Nemotron 3 Super 120B-A12B is part of the third generation of the Nemotron open model family, a series engineered specifically for building specialized, reliable AI agents rather than serving as a general-purpose chatbot. The "Super" designation marks a meaningful architectural step up from the lighter Nano variant, introducing several capabilities that simply weren't present before.
NVIDIA achieved this through their LatentMoE approach, where expert routing happens in a compressed latent space rather than full model dimensionality, meaning the system is smarter about which experts to engage, not just how many.
LatentMoE
Tokens are first projected into a compressed latent space for expert routing and computation. This reduces the byte cost per unit of model intelligence, you call 4 experts but pay roughly the compute of 1. It's what makes the 12B active parameter count feel much larger than it is.
Mamba-2 + Attention Hybrid
Rather than relying purely on attention, the model interleaves Mamba-2 state-space layers with selective attention blocks. Mamba-2 handles long-range context efficiently; attention handles local precision. The combination is faster than a dense transformer at long contexts.
Multi-Token Prediction (MTP)
Unlike models that predict one token at a time, Nemotron 3 Super uses MTP layers as a native speculative decoding mechanism. This is what drives the 167+ tokens/sec output speed, it's not just hardware, it's baked into the model weights.
NVFP4 Pretraining
The Super model is the first in its family pretrained at NVFP4 precision rather than using it only for post-training quantization. This allowed training on the full 25T+ token corpus more efficiently without sacrificing the quality typically expected of BF16-trained models.
1 Million Token Context
Not just a theoretical maximum, the model outperforms both GPT-OSS-120B and Qwen3.5-122B on the RULER benchmark at the full 1M token setting. This matters for agent workflows where conversation and tool-use history must stay in context across hundreds of steps.
Use cases
This model is documented as suited to agentic workflows, including collaborative agents and multi-agent applications.
It is also documented for high-volume workloads such as IT ticket automation.
It is documented for RAG systems, including retrieval-augmented generation and full-document retrieval workflows.
It is documented for complex instruction-following and long-context reasoning tasks.
Capabilities
- Function calling
- Streaming
- Structured outputs
- Reasoning
Nemotron 3 Super 120B-A12B benchmarks by 
Artificial Analysis
See how Nemotron 3 Super 120B-A12B compares in generation speed and latency. Data retrieved Nov 3, 2026.
Benchmark results
Compare 6 models available on AI/ML API across 5 independent benchmarks.
Humanity's Last Exam
Expert-written questions across academic subjects.
GPQA Diamond
Graduate-level science questions written to resist web search.
Terminal-Bench 4.0
Agentic tasks completed in a command-line environment.
AA-LCR
Reasoning across long documents of about 100k tokens.
SciCode
Python coding tasks drawn from scientific research.
Nemotron 3 Super 120B-A12B API pricing
| Usage | Price | Unit |
|---|---|---|
| Input | $0.0952 | 1M tokens |
| Output | $0.4761 | 1M tokens |
Prices from the AI/ML API price list, checked on Nov 4, 2026.
Nemotron 3 Super 120B-A12B vs other models: token prices
AI/ML API input and output rates for 5 models, per million tokens.
| Model | Input | Output |
|---|---|---|
 Nemotron 3 Super 120B-A12B | $0.0952 | $0.4761 |
| Nemotron 3.5 Content Safety | $0.2116 | $0.2116 |
| Nemotron 3 Nano 30B A3B | $0.0529 | $0.2116 |
| Nemotron 3.5 Super VL | $0.2645 | $1.59 |
| Nemotron 3 Ultra | $0.6348 | $3.81 |
How to use the Nemotron 3 Super 120B-A12B API
View docs
Build with your agent
Give your agent the model details, code and documentation.
Paste it into your coding agent to build the integration.
1Get your API key
Create API keyexport AIMLAPI_KEY="8f2b4c7d9a1e6b3f…"2Make your first request
curl 'https://api.aimlapi.com/v1/chat/completions' \
-H "Authorization: Bearer $AIMLAPI_KEY" \
-H 'Content-Type: application/json' \
-d '{
"model": "nvidia/nemotron-3-super-120b-a12b",
"messages": [
{
"role": "user",
"content": "Write a haiku about the ocean."
}
]
}'import os
import requests
headers = {
"Authorization": f"Bearer {os.environ['AIMLAPI_KEY']}",
"Content-Type": "application/json",
}
payload = {
"model": "nvidia/nemotron-3-super-120b-a12b",
"messages": [
{
"role": "user",
"content": "Write a haiku about the ocean."
}
]
}
response = requests.post("https://api.aimlapi.com/v1/chat/completions", headers=headers, json=payload, timeout=120)
response.raise_for_status()
print(response.json())const response = await fetch("https://api.aimlapi.com/v1/chat/completions", {
method: "POST",
headers: {
Authorization: `Bearer ${process.env.AIMLAPI_KEY}`,
"Content-Type": "application/json",
},
body: JSON.stringify({
"model": "nvidia/nemotron-3-super-120b-a12b",
"messages": [
{
"role": "user",
"content": "Write a haiku about the ocean."
}
]
}),
});
if (!response.ok) throw new Error(`HTTP ${response.status}`);
console.log(await response.json());Models similar to Nemotron 3 Super 120B-A12B
Related articles
Nemotron 3 Super 120B-A12B FAQ
What is the context window of Nemotron 3 Super 120B-A12B?
Nemotron 3 Super 120B-A12B has a 262,144 tokens context window and can return up to 262,144 tokens.
What inputs and outputs does Nemotron 3 Super 120B-A12B support?
Nemotron 3 Super 120B-A12B takes text as input and returns text.
What is the model ID for Nemotron 3 Super 120B-A12B?
Use nvidia/nemotron-3-super-120b-a12b as the model id. Requests go to https://api.aimlapi.com/v1/chat/completions.
When was Nemotron 3 Super 120B-A12B released?
Nemotron 3 Super 120B-A12B became available on April 30, 2026.
How much does Nemotron 3 Super 120B-A12B cost?
Nemotron 3 Super 120B-A12B costs $0.0952 input and $0.4761 output per 1M tokens on AI/ML API, billed as you go with no subscription.
Does Nemotron 3 Super 120B-A12B support streaming?
Yes, Nemotron 3 Super 120B-A12B can stream responses as they are generated.
Who created Nemotron 3 Super 120B-A12B?
Nemotron 3 Super 120B-A12B was built by NVIDIA.
Does Nemotron 3 Super 120B-A12B support tool or function calling?
Yes, it supports function calling along with parallel tool calls.
Can Nemotron 3 Super 120B-A12B return structured output formats?
Yes, structured output is listed as a supported feature and capability.