Nemotron 3 Ultra has a larger context window than GPT-2 (1.0M tokens vs 1K tokens).
| Spec | GPT-2 | Nemotron 3 Ultra |
|---|---|---|
| Input price | — | Free |
| Output price | — | Free |
| Context window | 1K tokens | 1.0M tokens |
| Weights | Open | Open |
| Free API (OpenRouter) | — | Yes (1 provider) |
| Released | Nov 2019 | Jun 2026 |
Not enough votes to call it. On the specs, Nemotron 3 Ultra has the edge: bigger model tier, newer, bigger context window.
No hosts listed on OpenRouter.
Per million tokens. Prices and uptime via OpenRouter, checked 9 Sep 2026.
Nemotron 3 Ultra uses 26.5x more hedging