ModelsNVIDIA
Nemotron 3 Ultra
Context
tokens
Released
on OpenRouter
Usage rank
of 28 tracked models
Token share
OpenRouter · Aug 20, 2026
Price per million tokens
Input $/MTok
OpenRouter · Aug 21, 2026
Output $/MTok
OpenRouter · Aug 21, 2026
Cache read $/MTok
OpenRouter
| Source | Tracked slug | In $/MTok | Out $/MTok | Cache read | Cache write | Context | Effective | Last Δ (in) |
|---|---|---|---|---|---|---|---|---|
| OpenRouter | nvidia/nemotron-3-ultra-550b-a55b | $0.6 | $3.60 | $0.2 | — | 512.3K | Aug 21, 2026 | — |
| models.dev | nvidia/nvidia/nemotron-3-ultra-550b-a55b | $0.5 | $2.50 | $0.15 | — | 1M | Aug 21, 2026 | 16.7% |
latest tracked price per source · Δ = most recent input-price move within that source
2 sources tracking
Input $/MTok
Output $/MTok
per-source step lines · $/MTok
history since Jul 8, 2026
Context & availability
| Provider | NVIDIA |
|---|---|
| Context window | 1M tokens |
| Listed on OpenRouter | Jun 4, 2026 |
| Tracked price sources | OpenRouter · models.dev |
| Hugging Face | nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16 |
| OpenRouter slugs | nvidia/nemotron-3-ultra-550b-a55b · nvidia/nemotron-3-ultra-550b-a55b-20260604 · nvidia/nemotron-3-ultra-550b-a55b:free |
source: OpenRouter model catalog + tracked price feeds
id: nemotron-3-ultra
Share of OpenRouter tokens
source: OpenRouter model rankings · share of tracked coding-token volume
6% on Aug 20, 2026
Benchmark results
| Benchmark | Configuration | Score | Date | Source |
|---|---|---|---|---|
| AA Agentic Index | nvidia/nemotron-3-ultra-550b-a55b:free | 27.4index | — | or-embedded-aa |
| AA Agentic Index | nvidia/nemotron-3-ultra-550b-a55b | 27.4index | — | or-embedded-aa |
| AA Agentic Index | nvidia/nemotron-3-ultra-550b-a55b:batch | 27.5index | — | or-embedded-aa |
| AA Coding Index | nvidia/nemotron-3-ultra-550b-a55b:free | 49.3index | — | or-embedded-aa |
| AA Coding Index | nvidia/nemotron-3-ultra-550b-a55b | 49.3index | — | or-embedded-aa |
| AA Coding Index | nvidia/nemotron-3-ultra-550b-a55b:batch | 49.3index | — | or-embedded-aa |
| AA Intelligence Index | nvidia/nemotron-3-ultra-550b-a55b:free | 37.8index | — | or-embedded-aa |
| AA Intelligence Index | nvidia/nemotron-3-ultra-550b-a55b | 37.8index | — | or-embedded-aa |
| AA Intelligence Index | nvidia/nemotron-3-ultra-550b-a55b:batch | 38.3index | — | or-embedded-aa |
scores as published by each benchmark; configurations differ — compare within a row's benchmark only
Similar models
| # | Model | Tokens/day | 7d Δ | Head-to-head |
|---|---|---|---|---|
| 7 | GLM 5.2Z.ai | 533B | 1.3% | vs Nemotron 3 Ultra |
| 2 | Hy3Tencent | 1.3T | 12.2% | vs Nemotron 3 Ultra |
| 1 | DeepSeek V4 FlashDeepSeek | 2.7T | 35.1% | vs Nemotron 3 Ultra |
| 9 | MiniMax M3MiniMax | 243.4B | 17.7% | vs Nemotron 3 Ultra |
ranked by OpenRouter usage-rank proximity
snapshot Aug 20, 2026
Nemotron 3 Ultra FAQ
- How much does Nemotron 3 Ultra cost per token?
- Nemotron 3 Ultra is priced at $0.6 per million input tokens and $3.60 per million output tokens, per OpenRouter as of Aug 21, 2026.
- What is Nemotron 3 Ultra's context window?
- Nemotron 3 Ultra supports a context window of up to 1M tokens, per NVIDIA.
- How widely used is Nemotron 3 Ultra?
- Nemotron 3 Ultra accounts for about 6% of tracked OpenRouter coding-token volume as of Aug 20, 2026.
- What are the alternatives to Nemotron 3 Ultra?
- By OpenRouter usage, comparable models include GLM 5.2, Hy3, and DeepSeek V4 Flash. Side-by-side comparisons are linked in the alternatives table.
Track Nemotron 3 Ultra
Email me when Nemotron 3 Ultra's price moves or it gets deprecated.