‹ PublicAI Index
The LLM benchmark aggregator.
K2 Horizon 375B A23B
Institute of Foundation Models · 375B · A23B
Strongest in Agents (#47 of 298, on 2 of its 6 boards), weakest in Core abilities (#126 of 201). Above par in 15 of 17 scopes.
§ 1 · Profile
What it is good at
Bars run from 50 — the average of the models each source lists — so right of the line is above par. The middle column is the gap to whoever leads that scope.
Agents56−10.3#47/2982/6
Knowledge work59.6−11.1#50/2082/2
Tool use60.1−13.9—/81✱0/1
Web research50.3−13.3—/36✱0/2
Workflow automation51−15—/0✱✱
Core abilities47.1−18.5#126/2011/4
General intelligence45.8−23.2#121/1811/3
Long context59.3−5.3—/0✱✱
Knowledge50.9−11.6—/197✱0/3
Factuality54.1−6.7—/0✱✱
Reasoning51.3−14.1—/195✱0/5
Science54.3−9.2—/122✱0/1
Expert reasoning53.5−12.9—/0✱✱
Coding52.5−13.6—/179✱0/6
Agentic coding51.6−15.1—/173✱0/5
Code generation55.3−10—/20✱0/1
Repository Q&A60.7−6—/0✱✱
§ 3 · Sources
Where the numbers come from
5 publications, 27 figures. Every one links to the page it was read from.
Artificial Analysis Intelligence Index 18
GDPval-AA 1365
AA-Briefcase 1292
GDPVal-AA 1441tau3-Banking 34%Toolathlon Verified 65.3%Automation Bench Public 25.3%Apex-Agents (pass@1) 24.8%MCPMark 67.7%BrowseComp 72.8%WildClawBench 50.9%Terminal-Bench 2.1 70.2%SciCode 42.7%SWE-Atlas-QnA 48.4%SWE Bench Pro 42.6%Humanity's Last Exam (without tools) 32%GPQA Diamond 87.3%CritPt 8.6%AA-LCR 76%AA-Omniscience Accuracy 23%AA-Omniscience Non-Hallucination 74.7%
GDPVal-AA (Elo) 1441Toolathlon Verified 65.3%Terminal-Bench 2.1 70.2%SWE-bench Pro (strict) 42.6%MCPMark 67.7%SWE-Atlas-QnA (strict) 48.4%
Badge
[](https://publicai.io/model-index/m/k2-horizon-375b-a23b)