Largest Nemotron 3 model for maximum open-weight reasoning and agent accuracy
Specification
most-agreed values
Context
203K
Max output
203K
Released
2026-06-04
Knowledge cutoff
—
Retires
—
Open weights
yes
Input
text
Output
text
Price
US dollars per million tokens · most-agreed
Input
$0.60
Output
$2.40
Cache read
$0.12
Available from 5 hosts
| Host | In $/M | Out $/M | Cache rd | Cache wr | Context | Output | Retires |
|---|---|---|---|---|---|---|---|
| deepinfra | $0.50 | $2.20 | $0.10 | — | 262K | — | — |
| wandb | $0.50 | $2.15 | $0.10 | — | 262K | 262K | — |
| baseten | $0.60 | $2.40 | $0.12 | — | 203K | 203K | — |
| venice | $0.625 | $3.125 | $0.1875 | — | 256K | 33K | — |
| wandb · wandb | $0.75 | $2.75 | $0.15 | — | 262K | — | — |
Price history
append-only observations · a listing writes a row only when its price moves
| Date | Host | Field | $/M |
|---|---|---|---|
| 2026-09-02 | wandb | Cache read | $0.15 |
| 2026-09-15 | wandb | Cache read | $0.10 |
| 2026-09-02 | wandb | Input | $0.75 |
| 2026-09-15 | wandb | Input | $0.50 |
| 2026-09-02 | wandb | Output | $2.75 |
| 2026-09-15 | wandb | Output | $2.15 |
More from NVIDIA
most-hosted first