Board/Anthropic/Claude Opus 4.8
A

Claude Opus 4.8

Anthropic·May 28, 2026N/A

May's Opus generation — described as a modest but tangible step up from 4.7, and still a benchmark fixture: Grok 4.5 and Qwen3.8 Max are routinely graded against its max-effort numbers.

Status
Generally available
License
Proprietary API
Context window
200K
Max output
64K
Input · $/M
$5
Output · $/M
$25
Cached input · $/M
$0.5
Blended · $/M
$10.00
Parameters
Undisclosed
Inputs
Text + images

Benchmarks

This model has no published scores on the benchmarks we track yet. Specs and pricing below the fold are still fully sourced.

Reasoning & knowledge

  • GPQA Diamond

    no public score

  • MMLU-Pro

    no public score

  • Humanity's Last Exam

    no public score

  • ARC-AGI-2

    no public score

Coding

  • SWE-bench Verified

    no public score

  • SWE-bench Pro

    no public score

  • LiveCodeBench

    no public score

Agents & computer use

  • Terminal-Bench 2

    no public score

  • OSWorld 2.0

    no public score

  • BrowseComp

    no public score

Math

  • AIME 2026

    no public score

Known strengths

  • Proven agent reliability
  • Widely benchmarked reference point

Primary sources

  • Anthropic announcement
Compare this model →

More from Anthropic

  • AClaude Opus 5Jul 24, 2026 · 200K ctx
  • AClaude Sonnet 5Jun 30, 2026 · 1M ctx
  • AClaude Fable 5Jun 9, 2026 · 1M ctx
  • AClaude Haiku 4.5Oct 15, 2025 · 200K ctx