Head to head
gpt-oss-120bvsQwen3 8B
For tasks, about a month.
Verdict · measured basis
gpt-oss-120b costs $0.0011 per solved task against
Qwen3 8B at $0.0043 — ▼ 4.1x cheaper.
Over 200 tasks that is $0.66 a month.
On output token price alone gpt-oss-120b looks 2.7x cheaper. The ranking holds, but the margin changes.
Verified pricing, pass rate and cost per solved task for gpt-oss-120b and Qwen3 8B| Figure | gpt-oss-120b | Qwen3 8B |
|---|
| Input $/Mtok | $0.037 | $0.117 |
|---|
| Output $/Mtok | $0.17 | $0.455 |
|---|
| Cached input $/Mtok | $0.037 | $0.117 |
|---|
| Pass rate | 60% | 57% |
|---|
| Cost per solved task | $0.0011measured | $0.0043measured |
|---|
| Cost per month | $0.21 | $0.87 |
|---|
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