LiteLLMModels

Nvidia Nemotron 3 Ultra 550B A55B API Pricing

As of October 10, 2026, Nvidia Nemotron 3 Ultra 550B A55B costs $0.50 per 1M input tokens and $2.15 per 1M output tokens from Weights & Biases, the lowest of 3 providers, with cache reads at $0.10 per 1M tokens. It has a 262,000-token context window and returns up to 202,800 tokens per response.

By NVIDIAModel ID wandb/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55BPrices updated Checked
Input
$0.50 / 1M tokens
Output
$2.15 / 1M tokens
Cache read
$0.10 / 1M tokens
Context window
262,000 tokens
Max output
202,800 tokens
Providers
3
Added to LiteLLM
August 27, 2026

Nvidia Nemotron 3 Ultra 550B A55B pricing by provider

3 listings across 3 providers. Token prices are in US dollars per 1M tokens. The highlighted row is the price quoted above.

ProviderModel name in LiteLLMInputOutputCache readContext
Weights & Biaseswandb/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B$0.50$2.15$0.10262K
Basetenbaseten/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B$0.60$2.40$0.12203K
DeepInfradeepinfra/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B$0.50$2.20$0.10256K

What Nvidia Nemotron 3 Ultra 550B A55B costs in practice

1,000 requests with 2,000 input and 500 output tokens each

Weights & Biases$2.075
DeepInfra$2.10
Baseten$2.40

100 long-document requests with 100,000 input and 2,000 output tokens each

Weights & Biases$5.43
DeepInfra$5.44
Baseten$6.48

Nvidia Nemotron 3 Ultra 550B A55B features

Use Nvidia Nemotron 3 Ultra 550B A55B with LiteLLM

Call Nvidia Nemotron 3 Ultra 550B A55B through the LiteLLM Python SDK, or put it behind the LiteLLM proxy and send OpenAI-format requests to /v1/chat/completions. LiteLLM tracks the cost of every request at these prices.

Python SDK
from litellm import completion

response = completion(
    model="wandb/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B",
    messages=[{"role": "user", "content": "Hello!"}],
)
Proxy config.yaml
model_list:
  - model_name: nvidia-nemotron-3-ultra-550b-a55b
    litellm_params:
      model: wandb/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B
      # credentials: see the provider's LiteLLM docs page
Call the proxy
curl http://0.0.0.0:4000/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk-1234" \
  -d '{
    "model": "nvidia-nemotron-3-ultra-550b-a55b",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Nvidia Nemotron 3 Ultra 550B A55B FAQ

How much does Nvidia Nemotron 3 Ultra 550B A55B cost?

Nvidia Nemotron 3 Ultra 550B A55B costs $0.50 per 1M input tokens and $2.15 per 1M output tokens from Weights & Biases, the lowest of 3 providers, with cache reads at $0.10 per 1M tokens.

What is the context window of Nvidia Nemotron 3 Ultra 550B A55B?

Nvidia Nemotron 3 Ultra 550B A55B has a 262,000-token context window and returns up to 202,800 tokens per response.

Which providers offer Nvidia Nemotron 3 Ultra 550B A55B?

Nvidia Nemotron 3 Ultra 550B A55B is available from 3 providers through LiteLLM: Weights & Biases, Baseten and DeepInfra. The lowest price is on Weights & Biases.

How do I call Nvidia Nemotron 3 Ultra 550B A55B with an OpenAI-compatible API?

Use the LiteLLM Python SDK with model="wandb/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B", or add the model to the LiteLLM proxy's config.yaml and send requests to /v1/chat/completions from any OpenAI client. LiteLLM tracks the cost of every request at the prices on this page.

What features does Nvidia Nemotron 3 Ultra 550B A55B support?

Nvidia Nemotron 3 Ultra 550B A55B supports function calling, structured output, reasoning and prompt caching. It does not support image input.

When did LiteLLM add Nvidia Nemotron 3 Ultra 550B A55B?

Nvidia Nemotron 3 Ultra 550B A55B was added to LiteLLM's model price file on August 27, 2026.

Sources

Prices come from LiteLLM's open-source model_prices_and_context_window.json, which LiteLLM uses to calculate the cost of every request. Provider pricing pages:

LiteLLM docs: Weights & Biases, Baseten, DeepInfra.

Provider pages: Weights & Biases, Baseten, DeepInfra.