LFM2.5-2.6B

Composite indexes

• Artificial Analysis Intelligence Index#110 of 142
  8independentIntelligence Index v4.3; reasoning; estimatedArtificial Analysis: LFM2.5-2.6B ↗

  Leader: Claude Opus 5.5 at 58

Coding

• SciCode#94 of 94
  14.4%independentreasoningArtificial Analysis: LFM2.5-2.6B ↗

  Leader: Claude Opus 5.5 at 66.9%

• Terminal-Bench 2.x#96 of 103
  4.5%independentTerminal-Bench 2.1; reasoningArtificial Analysis: LFM2.5-2.6B ↗

  Leader: Claude Fable 5.1 at 91.4%

Agents and tool use

• GDPval-AA#97 of 103
  -22independentreasoningArtificial Analysis: LFM2.5-2.6B ↗

  Leader: Claude Opus 5.5 at 1846

Reasoning and knowledge

• GPQA Diamond#119 of 142
  55.8%independentreasoningArtificial Analysis: LFM2.5-2.6B ↗

  Leader: GPT-6 Astra at 96.1%

• Humanity’s Last Exam#115 of 147
  6.2%independentno tools; reasoningArtificial Analysis: LFM2.5-2.6B ↗

  Leader: Claude Opus 5.5 at 61.4%

• CritPt#128 of 139
  0%independentreasoningArtificial Analysis: LFM2.5-2.6B ↗

  Leader: GPT-5.6 Sol at 32.3%

Vision and long context

• AA Long Context Reasoning#127 of 139
  5.7%independentreasoningArtificial Analysis: LFM2.5-2.6B ↗

  Leader: Kimi K3 at 88.7%

Other published results