Head to head
Nemotron 3 Nano 30B A3BvsLlama 3.1 8B Instruct
For tasks, about a month.
Verdict · measured basis
Llama 3.1 8B Instruct costs $0.0000 per solved task against
Nemotron 3 Nano 30B A3B at $0.0002 — ▼ 5.8x cheaper.
Over 200 tasks that is $0.04 a month.
On output token price alone Llama 3.1 8B Instruct looks 2.5x cheaper. The ranking holds, but the margin changes.
Verified pricing, pass rate and cost per solved task for Nemotron 3 Nano 30B A3B and Llama 3.1 8B Instruct| Figure | Nemotron 3 Nano 30B A3B | Llama 3.1 8B Instruct |
|---|
| Input $/Mtok | $0.05 | $0.05 |
|---|
| Output $/Mtok | $0.2 | $0.08 |
|---|
| Cached input $/Mtok | $0.03 | $0.025 |
|---|
| Pass rate | 81% | 42% |
|---|
| Cost per solved task | $0.0002measured | $0.0000measured |
|---|
| Cost per month | $0.04 | $0.01 |
|---|
Cost per month against tasks per month · drag the marker
Solid measured · dashed modeled · dotted stale. The gap is the verdict.
Solvency Bench (solvency.dev) — first-party measurement · verified 2026-08-26 · prices from each provider's pricing page; measured rows ignore the tier · Methodology · All models at these settings