
nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B
550B hybrid Mamba-Transformer MoE with 55B active params, latent MoE routing, multi-token prediction, and up to 1M token context
OpenAI-compatible chat completion. NVIDIA recommends temperature=1.0 and top_p=0.95; toggle reasoning via chat_template_kwargs.enable_thinking.
Input
from openai import OpenAI
client = OpenAI(
api_key="BASETEN_API_KEY",
base_url="https://inference.baseten.co/v1"
)
# NVIDIA recommends temperature=1.0 and top_p=0.95 for all tasks.
# Toggle reasoning on/off via chat_template_kwargs.enable_thinking.
response = client.chat.completions.create(
model="nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B",
messages=[
{
"role": "user",
"content": "Tell me a fun fact about hummingbirds."
}
],
stream=True,
stream_options={
"include_usage": True,
"continuous_usage_stats": True
},
top_p=0.95,
max_tokens=256,
temperature=1,
presence_penalty=0,
frequency_penalty=0
)
for chunk in response:
if chunk.choices and chunk.choices[0].delta.content is not None:
print(chunk.choices[0].delta.content, end="", flush=True)Output
{
"id": "143",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "[Model output here]",
"role": "assistant",
"audio": null,
"function_call": null,
"tool_calls": null
}
}
],
"created": 1741224586,
"model": "",
"object": "chat.completion",
"service_tier": null,
"system_fingerprint": null,
"usage": {
"completion_tokens": 145,
"prompt_tokens": 38,
"total_tokens": 183,
"completion_tokens_details": null,
"prompt_tokens_details": null
}
}