The honest view from inside technical leadership
LLM Leaderboard · Anthropic

Claude Opus 4.8

Benchmarks, pricing and rank for Claude Opus 4.8, with every score linked to the leaderboard that published it.

← All models

#2952.8CTO Index
Provider
Anthropic
Weights
Closed
License
Proprietary
Price $/M in / out
$5 / $25
Context window
1M
Benchmarks covered
5 of 5

Benchmark results

Terminal-Bench 4.1Agentic coding
23.6%
#31 of 41 tracked · source
WebDev ArenaCoding (human-judged)
1555
#42 of 44 tracked · source
ARC-AGI-2Reasoning
72.1%
#20 of 28 tracked · source
Text ArenaGeneral chat (human-judged)
1482
#20 of 40 tracked · source
ARC-AGI-3Interactive reasoning
1.5%
#15 of 19 tracked · source

Snapshot 22 November 2026 · Scores copied from each benchmark's own leaderboard · How the CTO Index works

Compare Claude Opus 4.8

Frequently asked questions

How good is Claude Opus 4.8?

Claude Opus 4.8 ranks #29 of 39 models on the Diary of a CTO Index with 52.8 points.

How much does Claude Opus 4.8 cost?

Claude Opus 4.8 lists at $5 / $25 per million input / output tokens.

What are Claude Opus 4.8's benchmark scores?

Terminal-Bench 4.1: 23.6%; WebDev Arena: 1555; ARC-AGI-2: 72.1%; Text Arena: 1482; ARC-AGI-3: 1.5%.

What is Claude Opus 4.8's context window?

Claude Opus 4.8 supports a context window of 1M tokens.