AI Battle > Rankings > K2 Horizon 375B A23B

K2 Horizon 375B A23B — Placements in 24 Rankings (2026)

K2 Horizon 375B A23B, published by MBZUAI Institute of Foundation Models, appears in 24 of the 129 datasets AI Battle tracks. Its strongest published placement is #2 of 20 in AA-Omniscience Hallucination Rate.

Last updated: . Every placement below comes from the named public dataset; AI Battle joins them by slug and changes nothing else.

All placements for K2 Horizon 375B A23B

RankingRankValue
AA-Omniscience Hallucination Rate#2 of 200.26
AA-Briefcase Tool Calls Breakdown, Avg per Task#8 of 2064.0
AA-Omniscience Index: Token Usage#12 of 20751342
Model Size: Total and Active Parameters#9 of 1423.0
Output Tokens per Intelligence Index Task#14 of 2015750
Output Tokens per Intelligence Index Task#14 of 2015750
Mean Turns per Task#14 of 2084.2
AA-LCR v1.1: Output Tokens per Task#14 of 2012.4
SciCode: Output Tokens per Task#14 of 20353
Humanity's Last Exam: Output Tokens per Task#15 of 20181
GDPval-AA v2: Average Turns per Task#15 of 2045.0
AA-Briefcase Elo#16 of 201298
AA-Briefcase Elo#16 of 201298
AA-Briefcase Output Tokens per Task#16 of 2051956
Artificial Analysis Finance & Accounting Index#17 of 2032.3
GDPval-AA v2: Output Tokens per Task#17 of 2025039
Artificial Analysis Intelligence Index by Open Weights / Proprietary#18 of 2030.8
GDPval-AA v2 Leaderboard#18 of 201400
Artificial Analysis Intelligence Index by Open Weights / Proprietary#18 of 2030.8
Artificial Analysis Intelligence Index#18 of 2030.8
Artificial Analysis Intelligence Index#18 of 2030.8
AA-Omniscience Index#19 of 20-2.97
AA-Omniscience Index#19 of 20-2.97
Context Window#20 of 20524288

Price vs intelligence

intelligence index →input price / 1M → $4.00
Each dot is a model, joined by AI Battle from two public datasets — the intelligence index ranking and the published input pricing ranking. Every published value is shown as-is, no averaging. The dashed line is the value frontier: no model above or to the right of it offers a lower input price at its intelligence level.

How to read this profile

Each row is a separate measurement in a separate dataset — the position reflects the metric of that ranking, and a higher number is not "better" for every dataset (times and prices are costs). AI Battle deliberately does not average placements into a single score: datasets differ in coverage, so a total would hide more than it shows.

Raw data

Machine-readable profile: /data/models/k2-horizon-375b-a23b.json.

Method

Placements, joins and this page are computed by AI Battle from public datasets — see the methodology for sources, limitations and corrections.