Qwen 3.5 122B A10B — benchmark results

Alibaba's open Qwen 3.5 122B A10B MoE (10B active). Provider: Alibaba. Released 2026-02-24. Access: Open.

Unified ELO 1650 ± 10, rank #365 of 1776 rated models, from 179 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
LLM Stats (MVBench)76.6Score (%)100
LLM Stats (VideoMME w/o sub.)83.9Score (%)100
MATH-MC Level 599.92Accuracy (%)100
LLM Stats (MMBench-V1.1)92.8Score (%)97.1
RewardBench 2 Safety94.67Accuracy (%)96.1
MATH-MC Level 499.58Accuracy (%)95.6
SEA-HELM70.92Mean Score (%)95.3
LLM Stats (MMStar)82.9Score (%)95.2
LLM Stats (OCRBench)92.1Score (%)95.2
RewardBench 2 Math90.71Accuracy (%)94.1
SnakeBench33TrueSkill Rating94
MathVision86.2Overall Accuracy (%)93.8

Interactive version: theaggregate.ai/model?slug=qwen-3-5-122b-a10b · How the rankings work · Data refreshed daily, snapshot 2026-07-22.