Nemotron 3 Ultra :free

nemotron-3-ultra-550b-a55b:freeby NVIDIAZDR (all rungs)Free

NVIDIA Nemotron 3 Ultra is an open frontier-reasoning and orchestration model from NVIDIA, with 55B active parameters out of 550B total (MoE). Built on a hybrid Transformer-Mamba mixture-of-experts architecture, it...

Compare Open in Playground

Context

512K

Max output

no data

Pricing

Fastest

60 tok/s

Released

Jun 2026

Promotion

FreeYour limits

Selected nemotron-3-ultra-550b-a55b:free. Requests stop when the free allowance is unavailable or exhausted. They never fall back to paid usage.

Only routes covered by a free allowance or a zero-price provider are eligible. Paid routes and credit overflow do not apply to this selection. View regular model routes and settings.

Supported parameters

Sampling

Temperature0-2Top-p0-1Top-kStop sequences

Tools & structure

ToolsParallel tool callsStructured output

Reasoning

Reasoningnone · low · medium · high · xhigh · maxDefaultmedium

Streaming

Limits

Max-tokens fieldmax_tokens

Sending an unsupported field? See error reference.

Fireworks docs

Quickstart

I want you to route my LLM calls for "nemotron-3-ultra-550b-a55b:free" through the Experiential gateway instead of
calling the provider directly. It speaks the OpenAI Chat Completions API, so this is a base-URL
and key swap. Please:

1. Point the client at https://api-pr-2136.preview.experientiallabs.ai/v1 as the base URL.
2. Authenticate with my Experiential API key from the EXPLABS_API_KEY environment variable. If
   it isn't set, stop and tell me to create one under Settings -> API Keys and export it.
3. Use the model id "nemotron-3-ultra-550b-a55b:free" exactly.
4. Update every place my code builds an LLM client for this model to use that base URL and key,
   leaving streaming and tool-calls as they are.
5. Make one test call and confirm the reply has zero cost. Stop if the free allowance is unavailable.

Tell me which files you changed.

Set EXPLABS_API_KEY to an organization API key before running your agent.

Benchmarks

Hugging Face

MMLU-ProHugging Face · Aug 202686.8%
GPQA DiamondHugging Face · Aug 202687.9%
SWE-bench VerifiedHugging Face · Aug 202669.5%
Humanity's Last ExamHugging Face · Aug 202626.1%

Models