anthropic

Claude Opus 5

current · frontier · measured cost · verified 2026-08-21

List price · USD per million tokens

Input

Output

$25

Cached input

$0.5

Context

Provider pricing page · verified 2026-08-21

Cost per solved task · by task tier

light

$0.10

moderate

$0.10

heavy

$0.10

Solvency Bench (solvency.dev) — first-party measurement · verified 2026-08-26 · measured basis: the benchmark observed this cost, so no tier assumption touches it and the three tiers are identical

Cache hit (read) rate shown. 5m cache write $6.25, 1h cache write $10. Full 1M context at standard rate. inference_geo=us applies 1.1x.

Benchmark rows

Where the pass rate comes from

Every row that names this model, with its harness and cost basis. Rows are harness + model combinations, not properties of the bare model.

Benchmark rows naming Claude Opus 5: benchmark, entry, pass rate, observed or usage-repriced attempt cost, cost basis and run date
Benchmark	Entry	Pass	$ / attempt	Basis	Run / read
Artificial Analysis Coding Agent Index v1.4	Claude Code - Opus 5 (xhigh)	68.0%	$8.17	measured by source	2026-08-26
Solvency Bench — single-turn suite (solvency-bench-v0)	claude-code - claude-opus-5	100.0%	$0.10	measured by solvency	2026-08-26

measured peers

Measured rows only, ranked as on the calculator. Monthly figures at 200 tasks. Lower is better.

measured · the benchmark ran the model and observed this cost

Compare

Side-by-side pricing and cost per solved task with another current model.