| Nemotron 3 Ultra | GPT-6 Astra | |
|---|---|---|
| Input, per 1M tokens | $0.60 | $10.00 |
| Output, per 1M tokens | $2.40 | $50.00 |
| Context window | 262K | 1.05M |
| Classification or routing (600 tokens in, 10 out), per 1,000 | $0.384 | $6.50 |
| Chat reply (2,000 in, 400 out), per 1,000 | $2.16 | $40.00 |
| Retrieval-augmented answer (8,000 in, 500 out), per 1,000 | $6.00 | $105 |
| Confidence signal | no | no |
| Tool calling | yes | yes |
Which one should carry your traffic?
If Nemotron 3 Ultra answers your requests as well as GPT-6 Astra, moving that traffic saves 95% on chat-shaped requests. Averages cannot tell you whether it does for your requests. Nemotron 3 Ultra does not expose a confidence signal through OpenRouter, so Gimbix would test it as a full replacement rather than a first answer.
Nemotron 3 Ultra profile · GPT-6 Astra profile · Measure it on your traffic