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.

SpecificationTrinity Large ThinkingArcee AIClaude Opus 5Anthropic
ReleasedApr 1, 2026Jul 24, 2026
Context window262K1M
Max output262K128K
Input / 1M$0.22$5
Output / 1M$0.85$25
Cost per task$0.08$2.25
Arena Elo1,1591,393
Serving providers25
Parameters398.6BUndisclosed
LicenceotherProprietary
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

  • Trinity Large Thinking
  • Claude Opus 5
  • Intelligence Index

    Trinity Large Thinking
    18.6
    Claude Opus 5
    63.1
  • Coding Index

    Trinity Large Thinking
    25.8
    Claude Opus 5
    78.0
  • Agentic Index

    Trinity Large Thinkingnot measured
    Claude Opus 5
    59.2
  • Output speed

    Trinity Large Thinking
    67 t/s
    Claude Opus 5
    87 t/s
  • Context window

    Trinity Large Thinking
    262K
    Claude Opus 5
    1M
  • Latency · lower is better

    Trinity Large Thinking
    744ms
    Claude Opus 5
    1.12s
  • Blended price / 1M · lower is better

    Trinity Large Thinking
    $0.378
    Claude Opus 5
    $10
  • Cost per task · lower is better

    Trinity Large Thinking
    $0.08
    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
MetricTrinity Large ThinkingClaude Opus 5
Intelligence Index18.663.1
Coding Index25.878.0
Agentic Index59.2
Output speed67 t/s87 t/s
Context window262K1M
Latency · lower is better744ms1.12s
Blended price / 1M · lower is better$0.378$10
Cost per task · lower is better$0.08$2.25

Evaluation scores

Percentage correct on a common 0–100% scale

  • Trinity Large Thinking
  • Claude Opus 5
  • GPQA Diamond

    Trinity Large Thinking
    75.2%
    Claude Opus 5
    93.2%
  • Humanity's Last Exam

    Trinity Large Thinking
    15.8%
    Claude Opus 5
    54.9%
  • SciCode

    Trinity Large Thinking
    36.1%
    Claude Opus 5
    55.7%
  • τ²-bench

    Trinity Large Thinking
    90.1%
    Claude Opus 5
    42.1%
  • Terminal-Bench Hard

    Trinity Large Thinking
    22.7%
    Claude Opus 5not measured
  • AA-LCR long context

    Trinity Large Thinking
    38.3%
    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
EvaluationTrinity Large ThinkingClaude Opus 5
GPQA Diamond75.2%93.2%
Humanity's Last Exam15.8%54.9%
SciCode36.1%55.7%
τ²-bench90.1%42.1%
Terminal-Bench Hard22.7%
LiveCodeBench
AA-LCR long context38.3%75.7%
AIME 2025

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