Ternary Bonsai 2.9 B vs Nemotron 3 Super 120 B A12 B: cost comparison
Prices updated 2026-10-28 · computed from live provider data
Short answer: Ternary Bonsai 2.9 B is cheaper on a
3:1 input/output blend — $0.092 per million tokens versus
$0.172, a 46% difference.
Both offer the same 262,144-token context window, so cost is the deciding factor.
Head to head
| Ternary Bonsai 2.9 B | Nemotron 3 Super 120 B A12 B | |
|---|---|---|
| Input, per 1M tokens | $0.080 | $0.080 |
| Output, per 1M tokens | $0.130 | $0.450 |
| Blended 3:1, per 1M | $0.092 | $0.172 |
| Context window | 262,144 | 262,144 |
| Vendor | prism-ml | nvidia |
Monthly cost at real volumes
| Workload (input / output) | Ternary Bonsai 2.9 B | Nemotron 3 Super 120 B A12 B |
|---|---|---|
| 1M in / 250K out | $0.113 | $0.193 |
| 10M in / 2.5M out | $1.12 | $1.93 |
| 100M in / 25M out | $11.25 | $19.25 |
Which should you pick?
Because the context windows are identical, the cheaper option is the better default. Choose Ternary Bonsai 2.9 B unless you are already standardised on nvidia tooling and want to avoid a second integration.