Qwen 3.5 35B A3B — benchmark results

Alibaba's open Qwen 3.5 35B A3B MoE (3B active). Provider: Alibaba. Released 2026-02-24. Access: Open.

Unified ELO 1626 ± 14, rank #420 of 1776 rated models, from 176 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
LLM Stats (AndroidWorld_SR)71.1Score (%)100
JudgeBench Reasoning96.94Accuracy (%)98
InvisibleBench10.5Hard Fail Rate (self-reported)95
LIBRA - MatreshkaNames *68.97Dataset Total Score (%)91.7
LIBRA - ruBABILongQA168.38Dataset Total Score (%)91.7
LIBRA - ruBABILongQA254.97Dataset Total Score (%)91.7
LIBRA - ruBABILongQA3 *32.6Dataset Total Score (%)91.7
LIBRA - ruSciFi51.47Dataset Total Score (%)91.7
LIBRA - ruSciPassageCount *21.89Dataset Total Score (%)91.7
StemBind39.3F Overall (self-reported)91.3
SAGE71.79OA (%) (self-reported)90.9
LLM Stats (MMStar)81.9Score (%)90.5

Interactive version: theaggregate.ai/model?slug=qwen-3-5-35b-a3b · How the rankings work · Data refreshed daily, snapshot 2026-07-22.