Head to head
vs MI For light moderate heavy
Verdict · measured basis
Phi 4 costs $0.0000 per solved task against
Mistral Small 3 at $0.0000 — ▼ 1.0x cheaper .
Over 200 tasks that is $0.00 a month.
On output token price alone Mistral Small 3 looks 1.8x 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 Mistral Small 3 and Phi 4 Figure Mistral Small 3 Phi 4 Input $/Mtok $0.05 $0.07 Output $/Mtok $0.08 $0.14 Cached input $/Mtok $0.05 $0.07 Pass rate 61% 67% Cost per solved task $0.0000 measured $0.0000 measured Cost per month $0.01 $0.01
Cost per month against tasks per month · drag the marker
Monthly cost against tasks per month Log-log lines for Phi 4 (measured, $0.0000 per solved task), Mistral Small 3 (measured, $0.0000 per solved task) from 10 to 100,000 tasks a month. Marker at 200 tasks. 10 100 1k 10k 100k $0.001 $0.01 $0.1 $1 TASKS / MONTH (log) → 200 tasks Phi 4 · MEASURED · $0.0000 per solved task $0.01 Phi 4 MEASURED Mistral Small 3 · MEASURED · $0.0000 per solved task $0.01 Mistral Small 3 MEASURED Monthly cost against tasks per month Log-log lines for Phi 4 (measured, $0.0000 per solved task), Mistral Small 3 (measured, $0.0000 per solved task) from 10 to 100,000 tasks a month. Marker at 200 tasks. 10 100 1k 10k 100k $0.001 $0.01 $0.1 $1 TASKS / MONTH (log) → 200 tasks Phi 4 · MEASURED · $0.0000 per solved task $0.01 Mistral Small 3 · MEASURED · $0.0000 per solved task $0.01 — Phi 4 · MEASURED — Mistral Small 3 · 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