Loading model details...
Fetching the latest models and pricing from the API.
NVIDIA Nemotron 3 Super 120B A12B
latestModel IDnvidia/NVIDIA-Nemotron-3-Super-120B-A12B-FP8
NVIDIA Nemotron-3-Super-120B-A12B is an open-weight LLM built for agentic reasoning and high-volume workloads. Using a hybrid LatentMoE architecture (Mamba-2 + MoE + Attention) with Multi-Token Prediction (MTP) and native NVFP4 pretraining on 25T tokens, it delivers up to 2.2x higher throughput than GPT-OSS-120B and 7.5x higher than Qwen3.5-122B. With a native 1M-token context window and configurable thinking mode, it is purpose-built for collaborative agents, long-context reasoning, and IT automation across 7 languages.
256K Tokens (up to 1M) ContextFast InferenceCodingFunction CallingMultilingual
Save 20%
$0.43$0.35
/ 1M input tokens
$1.29$1.04
/ 1M output tokens
Request (Non streaming)
from openai import OpenAI
# Initialize the OpenAI client with Qubrid base URL
client = OpenAI(
base_url="https://qubrid.com/v1",
api_key="QUBRID_API_KEY",
)
response = client.chat.completions.create(
model="nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-FP8",
messages=[
{
"role": "user",
"content": "Explain the main benefits of using a chat completion API for text generation."
}
],
max_tokens=16000,
temperature=1,
top_p=0.95,
stream=False
)
print(response.choices[0].message.content)from openai import OpenAI
# Initialize the OpenAI client with Qubrid base URL
client = OpenAI(
base_url="https://qubrid.com/v1",
api_key="QUBRID_API_KEY",
)
response = client.chat.completions.create(
model="nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-FP8",
messages=[
{
"role": "user",
"content": "Explain the main benefits of using a chat completion API for text generation."
}
],
max_tokens=16000,
temperature=1,
top_p=0.95,
stream=False
)
print(response.choices[0].message.content)Example response
A chat completion API provides a standard way to send conversational input and receive model-generated text in a single request. Key benefits include: • Interoperability: any client can use HTTP with JSON request and response bodies. • Flexibility: system prompts, user messages, and parameters such as temperature and max tokens are easy to configure. • Observability: responses typically include token usage fields for cost and performance tracking. A typical integration sends a POST request with the model name and messages array, then reads the assistant message from the first choice in the response.
Request (Non streaming)
from openai import OpenAI
# Initialize the OpenAI client with Qubrid base URL
client = OpenAI(
base_url="https://qubrid.com/v1",
api_key="QUBRID_API_KEY",
)
response = client.chat.completions.create(
model="nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-FP8",
messages=[
{
"role": "user",
"content": "Explain the main benefits of using a chat completion API for text generation."
}
],
max_tokens=16000,
temperature=1,
top_p=0.95,
stream=False
)
print(response.choices[0].message.content)from openai import OpenAI
# Initialize the OpenAI client with Qubrid base URL
client = OpenAI(
base_url="https://qubrid.com/v1",
api_key="QUBRID_API_KEY",
)
response = client.chat.completions.create(
model="nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-FP8",
messages=[
{
"role": "user",
"content": "Explain the main benefits of using a chat completion API for text generation."
}
],
max_tokens=16000,
temperature=1,
top_p=0.95,
stream=False
)
print(response.choices[0].message.content)Example response
A chat completion API provides a standard way to send conversational input and receive model-generated text in a single request. Key benefits include: • Interoperability: any client can use HTTP with JSON request and response bodies. • Flexibility: system prompts, user messages, and parameters such as temperature and max tokens are easy to configure. • Observability: responses typically include token usage fields for cost and performance tracking. A typical integration sends a POST request with the model name and messages array, then reads the assistant message from the first choice in the response.
Streaming supported • Function calling supported • See all examples in Playground Open in Playground for streaming, files, and all parameters.
Open in Playground