Skip to content
SOLVENCY
Request a demo

Account controls are loading.

Head to head

gpt-oss-20bvsLlama 3.2 1B Instruct

For tasks, about a month.

Verdict · measured basis

gpt-oss-20b costs $0.0003 per solved task against Llama 3.2 1B Instruct at $0.0003 — ▼ 1.2x cheaper. Over 200 tasks that is $0.01 a month.

On output token price alone gpt-oss-20b looks 1.5x cheaper. The ranking holds, but the margin changes.

Verified pricing, pass rate and cost per solved task for gpt-oss-20b and Llama 3.2 1B Instruct
Figuregpt-oss-20bLlama 3.2 1B Instruct
Input $/Mtok$0.03$0.027
Output $/Mtok$0.13$0.201
Cached input $/Mtok$0.03$0.027
Pass rate94%8%
Cost per solved task$0.0003measured$0.0003measured
Cost per month$0.05$0.06

Cost per month against tasks per month · drag the marker

Monthly cost against tasks per monthLog-log lines for gpt-oss-20b (measured, $0.0003 per solved task), Llama 3.2 1B Instruct (measured, $0.0003 per solved task) from 10 to 100,000 tasks a month. Marker at 200 tasks.101001k10k100k$0.01$0.1$1$10TASKS / MONTH (log) →200 tasksgpt-oss-20b · MEASURED · $0.0003 per solved task$0.05Llama 3.2 1B Instruct · MEASURED · $0.0003 per solved task$0.06— gpt-oss-20b · MEASURED— Llama 3.2 1B Instruct · 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