The Orchestrator › Models › NVIDIA
NVIDIA: Nemotron 3 Ultra
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...
Specifications
| Provider | NVIDIA |
| Model ID | nvidia/nemotron-3-ultra-550b-a55b |
| Context window | 262K tokens (262,144) |
| Modalities | text → text |
| Tokenizer | Other |
| Router sources | OpenRouter, TrustedRouter, Vercel AI Gateway |
| Released | 2026-06-04 |
| Tracked since | 2026-11-14 |
| Popularity | #28 by usage · peak 5.4% spend share in an OpenRouter category |
| Reasoning | On by default |
Pricing (per million tokens)
| Variant | Source | API ID | Input | Output | Context |
|---|---|---|---|---|---|
| standard | OpenRouter | nvidia/nemotron-3-ultra-550b-a55b | $0.6 | $2.40 | 262K |
| free | OpenRouter | nvidia/nemotron-3-ultra-550b-a55b:free | Free | Free | 1M |
| standard | TrustedRouter | nvidia/nemotron-3-ultra-550b-a55b | $0.528 | $2.27 | 262K |
| standard | Vercel AI Gateway | nvidia/nemotron-3-ultra-550b-a55b | $0.6 | $2.40 | 1M |
Recorded price history
| Date | Input / 1M | Output / 1M | Context |
|---|---|---|---|
| 2026-11-14 | $0.6 | $2.40 | 1M |
| 2026-11-14 | $0.528 | $2.27 | 262K |
| 2026-11-14 | $0.6 | $2.40 | 262K |
Capabilities
Agentic tool useStructured outputsReasoning mode256K+ context windowControllable reasoningNo content storage (attested)Zero data retentionNo training on inputs (attested)
Trust flags come from TrustedRouter's per-provider attestations.
Supported parameters
frequency_penalty, include_reasoning, logit_bias, logprobs, max_tokens, min_p, presence_penalty, reasoning, reasoning_effort, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_logprobs, top_p