GPT-5.4 nano

Composite indexes

• Artificial Analysis Intelligence Index#65 of 142
  21independentAA Intelligence Index v4.3.2, xhighArtificial Analysis: GPT-5.4 nano (xhigh) ↗

  Leader: Claude Opus 5.5 at 58

Coding

• Terminal-Bench 4.0#50 of 88
  0.5%independentTerminal-Bench 4.0 (AA harness), xhighArtificial Analysis: GPT-5.4 nano (xhigh) ↗

  Leader: Claude Mythos 5.1 at 60.9%

• Terminal-Bench 2.x#47 of 103
  60.7%independentTerminal-Bench 2.1 (AA harness), xhighArtificial Analysis: GPT-5.4 nano (xhigh) ↗

  Leader: Claude Fable 5.1 at 91.4%

• SciCode#49 of 94
  47.2%independentxhighArtificial Analysis: GPT-5.4 nano (xhigh) ↗

  Leader: Claude Opus 5.5 at 66.9%

Agents and tool use

• GDPval-AA#58 of 103
  937independentGDPval-AA, xhighArtificial Analysis: GPT-5.4 nano (xhigh) ↗

  Leader: Claude Opus 5.5 at 1846

• τ²-Bench#34 of 88
  76%independenttelecom (AA), xhighArtificial Analysis: GPT-5.4 nano (xhigh) ↗

  Leader: GLM-5.2 at 99.1%

Reasoning and knowledge

• Humanity’s Last Exam#59 of 147
  28.3%independentno tools (AA), xhighArtificial Analysis: GPT-5.4 nano (xhigh) ↗

  Leader: Claude Opus 5.5 at 61.4%

• GPQA Diamond#68 of 142
  81.7%independentAA, xhighArtificial Analysis: GPT-5.4 nano (xhigh) ↗

  Leader: GPT-6 Astra at 96.1%

• CritPt#42 of 139
  9.3%independentxhighArtificial Analysis: GPT-5.4 nano (xhigh) ↗

  Leader: GPT-5.6 Sol at 32.3%

• ARC-AGI-2#22 of 25
  5.7%independentsemi-private, XHighARC Prize leaderboard ↗

  Leader: GPT-6 Astra at 95%

Vision and long context

• AA Long Context Reasoning#53 of 139
  76.7%independentxhighArtificial Analysis: GPT-5.4 nano (xhigh) ↗

  Leader: Kimi K3 at 88.7%

• MMMU-Pro#48 of 72
  65.4%independentAA, xhighArtificial Analysis: GPT-5.4 nano (xhigh) ↗

  Leader: Claude Opus 5.5 at 87.7%

Human preference

• LMArena Text#47 of 54
  1402independentgpt-5.4-nano-high, style control default, vote cutoff 2026-09-13LMArena Text leaderboard ↗

  Leader: Claude Fable 5 at 1506

Other published results