Head to head
vs BY For light moderate heavy
Verdict · measured basis
UI-TARS 7B costs $0.0003 per solved task against
gpt-oss-120b at $0.0011 — ▼ 4.0x cheaper .
Over 200 tasks that is $0.16 a month.
On output token price alone gpt-oss-120b looks 1.2x cheaper. The ranking reverses once you divide by pass rate — the cheaper tokens do not win the task.
Copy link Save scenario Export compare table Planned
Verified pricing, pass rate and cost per solved task for gpt-oss-120b and UI-TARS 7B Figure gpt-oss-120b UI-TARS 7B Input $/Mtok $0.037 $0.1 Output $/Mtok $0.17 $0.2 Cached input $/Mtok $0.037 $0.1 Pass rate 60% 25% Cost per solved task $0.0011 measured $0.0003 measured Cost per month $0.21 $0.05
Cost per month against tasks per month · drag the marker
Monthly cost against tasks per month Log-log lines for UI-TARS 7B (measured, $0.0003 per solved task), gpt-oss-120b (measured, $0.0011 per solved task) from 10 to 100,000 tasks a month. Marker at 200 tasks. 10 100 1k 10k 100k $0.01 $0.1 $1 $10 $100 TASKS / MONTH (log) → 200 tasks UI-TARS 7B · MEASURED · $0.0003 per solved task $0.05 UI-TARS 7B MEASURED gpt-oss-120b · MEASURED · $0.0011 per solved task $0.21 gpt-oss-120b MEASURED Monthly cost against tasks per month Log-log lines for UI-TARS 7B (measured, $0.0003 per solved task), gpt-oss-120b (measured, $0.0011 per solved task) from 10 to 100,000 tasks a month. Marker at 200 tasks. 10 100 1k 10k 100k $0.01 $0.1 $1 $10 $100 TASKS / MONTH (log) → 200 tasks UI-TARS 7B · MEASURED · $0.0003 per solved task $0.05 gpt-oss-120b · MEASURED · $0.0011 per solved task $0.21 — UI-TARS 7B · MEASURED — gpt-oss-120b · MEASURED 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