LongCat-2.0

Composite indexes

• Artificial Analysis Intelligence Index#67 of 142
  19independentv4.3.2Artificial Analysis: LongCat 2.0 ↗

  Leader: Claude Opus 5.5 at 58

Coding

• Terminal-Bench 4.0#72 of 88
  0%independentAAArtificial Analysis: LongCat 2.0 ↗

  Leader: Claude Mythos 5.1 at 60.9%

• Terminal-Bench 2.x#62 of 103
  70.8%vendorTerminal-Bench 2.1Hugging Face: meituan-longcat/LongCat-2.0 ↗
  50.2%independentAA, Terminal-Bench 2.1Artificial Analysis: LongCat 2.0 ↗

  Leader: Claude Fable 5.1 at 91.4%

• SciCode#80 of 94
  36.3%independentAAArtificial Analysis: LongCat 2.0 ↗

  Leader: Claude Opus 5.5 at 66.9%

• SWE-bench Pro#13 of 31
  59.5%vendorHugging Face: meituan-longcat/LongCat-2.0 ↗

  Leader: Claude Opus 5.5 at 89.9%

Agents and tool use

• GDPval-AA#61 of 103
  856independentGDPval-AA v2.1Artificial Analysis: LongCat 2.0 ↗

  Leader: Claude Opus 5.5 at 1846

• BrowseComp#6 of 14
  79.9%vendorHugging Face: meituan-longcat/LongCat-2.0 ↗

  Leader: GPT-6 Astra at 91.5%

Reasoning and knowledge

• Humanity’s Last Exam#51 of 147
  33.7%independentAA, no toolsArtificial Analysis: LongCat 2.0 ↗

  Leader: Claude Opus 5.5 at 61.4%

• GPQA Diamond#76 of 142
  88.9%vendorHugging Face: meituan-longcat/LongCat-2.0 ↗
  78%independentAAArtificial Analysis: LongCat 2.0 ↗

  Leader: GPT-6 Astra at 96.1%

• CritPt#60 of 139
  2.6%independentAAArtificial Analysis: LongCat 2.0 ↗

  Leader: GPT-5.6 Sol at 32.3%

Vision and long context

• AA Long Context Reasoning#79 of 139
  65%independentAAArtificial Analysis: LongCat 2.0 ↗

  Leader: Kimi K3 at 88.7%

Other published results