Head to head
gpt-oss-20bvsLlama 3 8B Lunaris
For tasks, about a month.
Verdict · measured basis
Llama 3 8B Lunaris costs $0.0001 per solved task against
gpt-oss-20b at $0.0003 — ▼ 4.4x cheaper.
Over 200 tasks that is $0.04 a month.
On output token price alone Llama 3 8B Lunaris looks 2.6x cheaper. The ranking holds, but the margin changes.
Verified pricing, pass rate and cost per solved task for gpt-oss-20b and Llama 3 8B Lunaris| Figure | gpt-oss-20b | Llama 3 8B Lunaris |
|---|
| Input $/Mtok | $0.03 | $0.04 |
|---|
| Output $/Mtok | $0.13 | $0.05 |
|---|
| Cached input $/Mtok | $0.03 | $0.04 |
|---|
| Pass rate | 94% | 17% |
|---|
| Cost per solved task | $0.0003measured | $0.0001measured |
|---|
| Cost per month | $0.05 | $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