Kimi Linear 48B A3B Instruct: benchmark results

Moonshot's 48B MoE (3B active) with hybrid Kimi Delta linear attention and a 1M context, MIT-licensed. Provider: Moonshot. Released 2025-10-30. Access: Open.

Unified ELO 1495 ± 1, rank #714 of 1392 rated models, from 21 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
UGI - Writing37.74Writing Score64.8
UGI - Natural Intelligence22.09NatInt Score51.2
AA Terminal-Bench Hard11.36Accuracy (%)46.8
SpeechMap Compliance53.4% Requests Completed37.2
AA AIME 202536.33Accuracy (%)33.7
AA LiveCodeBench37.78Pass@1 (%)33.3
AA Long Context Reasoning28Accuracy (%)30.9
BenchmarkList ECI94.51Capability Index (ECI)27.2
Artificial Analysis Intelligence Index2.9Intelligence Index26.4
ATLAS33.1ATLAScore (self-reported)22.7
UGI - Willingness (W/10)2.8W/10 Score20.6
SvelteBench66.7Average pass@1 (%)19.2

Interactive version: theaggregate.ai/model?slug=kimi-linear-48b-a3b-instruct · How It Works · Data refreshed daily, snapshot 2026-09-05.