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.

SpecificationGLM 5.2Z.aiClaude Opus 5Anthropic
ReleasedJun 16, 2026Jul 24, 2026
Context window1.05M1M
Max output262K128K
Input / 1M$0.76$5
Output / 1M$2.42$25
Cost per task$0.23$2.25
Arena Elo1,3631,393
Serving providers275
Parameters753.3BUndisclosed
LicencemitProprietary
Capabilities
  • Reasoning
  • Tool use
  • Open weights
  • 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

  • GLM 5.2
  • Claude Opus 5

  • Intelligence Index

    GLM 5.2
    52.6
    Claude Opus 5
    63.1

  • Coding Index

    GLM 5.2
    68.8
    Claude Opus 5
    78.0

  • Agentic Index

    GLM 5.2
    45.7
    Claude Opus 5
    59.2

  • Output speed

    GLM 5.2
    142 t/s
    Claude Opus 5
    87 t/s

  • Context window

    GLM 5.2
    1.05M
    Claude Opus 5
    1M

  • Latency · lower is better

    GLM 5.2
    712ms
    Claude Opus 5
    1.12s

  • Blended price / 1M · lower is better

    GLM 5.2
    $1.18
    Claude Opus 5
    $10

  • Cost per task · lower is better

    GLM 5.2
    $0.23
    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
MetricGLM 5.2Claude Opus 5
Intelligence Index52.663.1
Coding Index68.878.0
Agentic Index45.759.2
Output speed142 t/s87 t/s
Context window1.05M1M
Latency · lower is better712ms1.12s
Blended price / 1M · lower is better$1.18$10
Cost per task · lower is better$0.23$2.25

Evaluation scores

Percentage correct on a common 0–100% scale

  • GLM 5.2
  • Claude Opus 5

  • GPQA Diamond

    GLM 5.2
    89.5%
    Claude Opus 5
    93.2%

  • Humanity's Last Exam

    GLM 5.2
    41.1%
    Claude Opus 5
    54.9%

  • SciCode

    GLM 5.2
    50.5%
    Claude Opus 5
    55.7%

  • τ²-bench

    GLM 5.2
    99.1%
    Claude Opus 5
    42.1%

  • Terminal-Bench Hard

    GLM 5.2
    50.8%
    Claude Opus 5not measured

  • AA-LCR long context

    GLM 5.2
    76.7%
    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
EvaluationGLM 5.2Claude Opus 5
GPQA Diamond89.5%93.2%
Humanity's Last Exam41.1%54.9%
SciCode50.5%55.7%
τ²-bench99.1%42.1%
Terminal-Bench Hard50.8%—
LiveCodeBench——
AA-LCR long context76.7%75.7%
AIME 2025——

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