RankingIA v0.6.1
EN

Qwen3.5-Max-Preview

Alibaba

No score: no objective benchmark
Input / output
—
US$ per 1M tokens
Context
—
tokens
Release
—
release date
Weights
Closed
 

Family history

The 7 versions of this family, from first to newest, by release date. Click a point to open that version.

See the data as a table
ModelReleasedScoreRankAA Intel.Elo Arena
Qwen-Max————1282
Qwen2.5 Max01/28/202531.5#337—1367
Qwen3 Max Thinking01/26/202666.5#56—1439
Qwen3.5-Max-Preview————1470
Qwen3.6 Max Preview04/20/202667.1#53—1447
Qwen3.7 Max05/21/202672.5#33—1475
Qwen3.8 Max (0902)09/03/202668.3#4833.61483

Scores by source

Scores by sourcepercentile = % of ranked models it beats
MetricValuePercentileSource
Arena text (Elo) weight 2 1470 86% LMArena

Compare with other models

At the sources