Skip to content
SOLVENCY

anthropic

Claude Opus 4.6

legacy · frontier · modelled cost · verified 2026-08-21

List price · USD per million tokens

Input

$5

Output

$25

Cached input

$0.5

Context

1M

Provider pricing page · verified 2026-08-21

Cost per solved task · by task tier

light

$0.24

moderate

$3.85

heavy

$8.67

Source: Scale SEAL leaderboard, SWE-bench Pro · verified 2026-08-21 · modelled basis: pass rate published, cost from Solvency's loop model at current prices

Listed under legacy models on the models overview page; still list-priced.

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 4.6: benchmark, entry, pass rate, measured cost per task, cost basis and run date
BenchmarkEntryPass$ / taskBasisRun / read
Scale SEAL leaderboard - SWE-bench Pro (public)claude-opus-4-6 (thinking)51.9%not publishedmodelled by solvency2026-08-21

modelled peers

Cost per solved task against its peers

Modelled rows only, heavy tier, 200 tasks a month. Lower is better. These are Solvency estimates, not observed costs.

modelled · pass rate published; cost is Solvency's loop model at verified prices

MODELLED: cost per solved task, ranked4 modelled rows ranked cheapest first. 1 GPT-5.4 $4.19; 2 Gemini 3.1 Pro (preview) $4.30; 3 Claude Opus 4.6 $8.67; 4 Claude Opus 4.5 $9.81. Monthly figures at 200 tasks.pass rate 59%GPT-5.4$4.19vs ›$838/mopass rate 46%Gemini 3.1 Pro (preview)$4.30vs ›$859/mopass rate 52%Claude Opus 4.6$8.67$1.7k/mopass rate 46%Claude Opus 4.5$9.81vs ›$2.0k/mo

Source: Scale SEAL leaderboard, SWE-bench Pro · verified 2026-08-21

Compare

Head to head

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