Llama 4 Scout

Composite indexes

• Artificial Analysis Intelligence Index#111 of 142
  8independentAA Intelligence Index v4.3.2, non-reasoningArtificial Analysis: Llama 4 Scout ↗

  Leader: Claude Opus 5.5 at 58

Coding

• Terminal-Bench 4.0#64 of 88
  0%independentTerminal-Bench 4.0 (AA harness), non-reasoningArtificial Analysis: Llama 4 Scout ↗

  Leader: Claude Mythos 5.1 at 60.9%

• Terminal-Bench 2.x#98 of 103
  3.7%independentTerminal-Bench 2.1 (AA harness), non-reasoningArtificial Analysis: Llama 4 Scout ↗

  Leader: Claude Fable 5.1 at 91.4%

• SciCode#91 of 94
  21.3%independentnon-reasoningArtificial Analysis: Llama 4 Scout ↗

  Leader: Claude Opus 5.5 at 66.9%

• LiveCodeBench#43 of 56
  29.9%independentAA, non-reasoningArtificial Analysis: Llama 4 Scout ↗

  Leader: Fugu at 92.9%

Agents and tool use

• GDPval-AA#100 of 103
  -178independentGDPval-AA, non-reasoningArtificial Analysis: Llama 4 Scout ↗

  Leader: Claude Opus 5.5 at 1846

• τ²-Bench#73 of 88
  15.5%independenttelecom (AA), non-reasoningArtificial Analysis: Llama 4 Scout ↗

  Leader: GLM-5.2 at 99.1%

Reasoning and knowledge

• Humanity’s Last Exam#140 of 147
  3.8%independentno tools (AA), non-reasoningArtificial Analysis: Llama 4 Scout ↗

  Leader: Claude Opus 5.5 at 61.4%

• GPQA Diamond#112 of 142
  58.7%independentAA, non-reasoningArtificial Analysis: Llama 4 Scout ↗

  Leader: GPT-6 Astra at 96.1%

• CritPt#92 of 139
  0%independentnon-reasoningArtificial Analysis: Llama 4 Scout ↗

  Leader: GPT-5.6 Sol at 32.3%

• ARC-AGI-2#25 of 25
  0%independentsemi-private,ARC Prize leaderboard ↗

  Leader: GPT-6 Astra at 95%

Vision and long context

• AA Long Context Reasoning#107 of 139
  27.7%independentnon-reasoningArtificial Analysis: Llama 4 Scout ↗

  Leader: Kimi K3 at 88.7%

• MMMU-Pro#63 of 72
  52.9%independentAA, non-reasoningArtificial Analysis: Llama 4 Scout ↗

  Leader: Claude Opus 5.5 at 87.7%

Human preference

• LMArena Text#53 of 54
  1321independentllama-4-scout, style control default, vote cutoff 2026-09-13LMArena Text leaderboard ↗

  Leader: Claude Fable 5 at 1506

Other published results