Skip to content
llmwaves

Compare

Four models, one page

Pick the models you are actually choosing between. Every row is scaled within itself, so a bar means something next to its neighbours and nothing across rows.

SpecificationClaude Opus 4.8AnthropicClaude Opus 5Anthropic
ReleasedMay 27, 2026Jul 24, 2026
Context window1M1M
Max output128K128K
Input / 1M$5$5
Output / 1M$25$25
Cost per task$2.25$2.25
Arena Elo1,2851,393
Serving providers45
ParametersUndisclosedUndisclosed
LicenceProprietaryProprietary
Capabilities
  • Reasoning
  • Tool use
  • Vision
  • Reasoning
  • Tool use
  • Vision

Metrics side by side

Each row is scaled to the largest value in that row — bars compare within a row, never across rows

  • Claude Opus 4.8
  • Claude Opus 5

  • Intelligence Index

    Claude Opus 4.8
    57.3
    Claude Opus 5
    63.1

  • Coding Index

    Claude Opus 4.8
    74.3
    Claude Opus 5
    78.0

  • Agentic Index

    Claude Opus 4.8
    49.4
    Claude Opus 5
    59.2

  • Output speed

    Claude Opus 4.8
    64 t/s
    Claude Opus 5
    87 t/s

  • Context window

    Claude Opus 4.8
    1M
    Claude Opus 5
    1M

  • Latency · lower is better

    Claude Opus 4.8
    902ms
    Claude Opus 5
    1.12s

  • Blended price / 1M · lower is better

    Claude Opus 4.8
    $10
    Claude Opus 5
    $10

  • Cost per task · lower is better

    Claude Opus 4.8
    $2.25
    Claude Opus 5
    $2.25

Rows marked “lower is better” still draw a longer bar for a larger number — read the value, not just the length. Arena Elo is in the specification table above instead: it has no meaningful zero, so a bar would flatten the gaps.

View as table
MetricClaude Opus 4.8Claude Opus 5
Intelligence Index57.363.1
Coding Index74.378.0
Agentic Index49.459.2
Output speed64 t/s87 t/s
Context window1M1M
Latency · lower is better902ms1.12s
Blended price / 1M · lower is better$10$10
Cost per task · lower is better$2.25$2.25

Evaluation scores

Percentage correct on a common 0–100% scale

  • Claude Opus 4.8
  • Claude Opus 5

  • GPQA Diamond

    Claude Opus 4.8
    92.0%
    Claude Opus 5
    93.2%

  • Humanity's Last Exam

    Claude Opus 4.8
    48.7%
    Claude Opus 5
    54.9%

  • SciCode

    Claude Opus 4.8
    53.5%
    Claude Opus 5
    55.7%

  • τ²-bench

    Claude Opus 4.8
    94.4%
    Claude Opus 5
    42.1%

  • Terminal-Bench Hard

    Claude Opus 4.8
    58.3%
    Claude Opus 5not measured

  • AA-LCR long context

    Claude Opus 4.8
    73.0%
    Claude Opus 5
    75.7%

A missing bar means that evaluation was not run for that model — it is not a zero.

View as table
EvaluationClaude Opus 4.8Claude Opus 5
GPQA Diamond92.0%93.2%
Humanity's Last Exam48.7%54.9%
SciCode53.5%55.7%
τ²-bench94.4%42.1%
Terminal-Bench Hard58.3%—
LiveCodeBench——
AA-LCR long context73.0%75.7%
AIME 2025——

Estimated from list pricing: 50K input tokens plus 80K output tokens for reasoning models (25K for non-reasoning).