Trinity Large Thinking

Composite indexes

• Artificial Analysis Intelligence Index#96 of 142
  11independentIntelligence Index v4.3; reasoningArtificial Analysis: Trinity Large Thinking ↗

  Leader: Claude Opus 5.5 at 58

Coding

• SciCode#67 of 94
  40.6%independentreasoningArtificial Analysis: Trinity Large Thinking ↗

  Leader: Claude Opus 5.5 at 66.9%

• Terminal-Bench 4.0#59 of 88
  0.5%independentreasoningArtificial Analysis: Trinity Large Thinking ↗

  Leader: Claude Mythos 5.1 at 60.9%

• Terminal-Bench 2.x#84 of 103
  20.6%independentTerminal-Bench 2.1; reasoningArtificial Analysis: Trinity Large Thinking ↗

  Leader: Claude Fable 5.1 at 91.4%

Agents and tool use

• τ²-Bench#20 of 88
  90.1%independenttelecom; reasoningArtificial Analysis: Trinity Large Thinking ↗

  Leader: GLM-5.2 at 99.1%

• GDPval-AA#86 of 103
  324independentreasoningArtificial Analysis: Trinity Large Thinking ↗

  Leader: Claude Opus 5.5 at 1846

Reasoning and knowledge

• GPQA Diamond#89 of 142
  75.2%independentreasoningArtificial Analysis: Trinity Large Thinking ↗

  Leader: GPT-6 Astra at 96.1%

• Humanity’s Last Exam#79 of 147
  15.8%independentno tools; reasoningArtificial Analysis: Trinity Large Thinking ↗

  Leader: Claude Opus 5.5 at 61.4%

• CritPt#73 of 139
  0.9%independentreasoningArtificial Analysis: Trinity Large Thinking ↗

  Leader: GPT-5.6 Sol at 32.3%

Vision and long context

• AA Long Context Reasoning#99 of 139
  38%independentreasoningArtificial Analysis: Trinity Large Thinking ↗

  Leader: Kimi K3 at 88.7%

Other published results