Grok 4.3

Composite indexes

• Artificial Analysis Intelligence Index#64 of 151
  25independentAA Intelligence Index v4.3.2, highArtificial Analysis: Grok 4.3 (high) ↗

  Leader: Claude Opus 5.5 at 58

Coding

• Terminal-Bench 4.0#70 of 96
  0%independentTerminal-Bench 4.0 (AA harness), highArtificial Analysis: Grok 4.3 (high) ↗

  Leader: Claude Sonnet 5.5 at 70.6%

• Terminal-Bench 2.x#71 of 105
  39.7%independentTerminal-Bench 2.1 (AA harness), highArtificial Analysis: Grok 4.3 (high) ↗

  Leader: Claude Fable 5.1 at 91.4%

• SciCode#51 of 100
  48.3%independenthighArtificial Analysis: Grok 4.3 (high) ↗

  Leader: Claude Opus 5.5 at 66.9%

Agents and tool use

• GDPval-AA#65 of 109
  923independentGDPval-AA, highArtificial Analysis: Grok 4.3 (high) ↗

  Leader: Muse Wonder Full at 1857

• τ²-Bench#5 of 89
  97.7%independenttelecom (AA), highArtificial Analysis: Grok 4.3 (high) ↗

  Leader: GLM-5.2 at 99.1%

Reasoning and knowledge

• Humanity’s Last Exam#47 of 153
  37.2%independentno tools (AA), highArtificial Analysis: Grok 4.3 (high) ↗

  Leader: Claude Sonnet 5.5 at 64.5%

• GPQA Diamond#39 of 144
  90.1%independentAA, highArtificial Analysis: Grok 4.3 (high) ↗

  Leader: GPT-6 Astra at 96.1%

• CritPt#51 of 144
  8%independenthighArtificial Analysis: Grok 4.3 (high) ↗

  Leader: Muse Wonder Full at 33%

Vision and long context

• AA Long Context Reasoning#69 of 144
  73%independenthighArtificial Analysis: Grok 4.3 (high) ↗

  Leader: Kimi K3 at 88.7%

• MMMU-Pro#27 of 72
  78.1%independentAA, highArtificial Analysis: Grok 4.3 (high) ↗

  Leader: Claude Opus 5.5 at 87.7%

Human preference

• LMArena Text#39 of 57
  1443independentgrok-4.3, style control default, vote cutoff 2026-09-13LMArena Text leaderboard ↗

  Leader: Gemini 4 Argon at 1525

• LMArena WebDev#55 of 58
  1357independentgrok-4.3, LMArena Code arena (formerly WebDev), vote cutoff 2026-09-23LMArena Code (WebDev) leaderboard ↗

  Leader: Claude Opus 5.5 at 1818

Other published results