Anthropic
nemotron-3-ultra-550b-a55b:batch
nvidia/nemotron-3-ultra-550b-a55b:batch
Price per million tokens
- Input
- $0.760
- Output
- $4.558
- Cache write
- not offered
- Cache read
- $0.253
What you pay. It is what the tokens cost us to buy plus 20%, and your usage page shows what we paid beside what you paid on every request you send.
Limits
- Context window
- 512,288 tokens
- Max output
- provider default
- Images
- text only
What it costs to run
| Workload | Input | Output | Cost | Per 1,000 calls |
|---|---|---|---|---|
| Short chat turn | 1,000 | 500 | $0.0030 | $3.04 |
| Long document summary | 50,000 | 2,000 | $0.0471 | $47.10 |
| Agent step with tools | 20,000 | 1,500 | $0.0220 | $22.03 |
| 1M in, 1M out | 1,000,000 | 1,000,000 | $5.3172 | $5317.20 |
Calling it
const res = await client.chat.completions.create({ model: "nvidia/nemotron-3-ultra-550b-a55b:batch", messages: [{ role: "user", content: "Hello" }], });