Alibaba 🇨🇳 · Qwen3.8

Qwen3.8-Max-0902

Dated September 2 snapshot of Alibaba's Qwen3.8-Max, available in Model Studio for text, reasoning, visual understanding, coding and long-horizon agent workflows. Artificial Analysis measures its Intelligence Index at 45 against 40 for the earlier August snapshot. The reviewed source does not establish artifact access or a proprietary licensing status for this exact snapshot.

DOOM SCORE87.9out of 100model risk profile, not the overall index
0 comments · 0 votesOpen discussion

Public discussion is readable by everyone. Sign in to comment, reply, or vote.

No comments yet. Start the discussion.

CURRENT ASSESSMENT · REVISION 1

Why this model scores 87.9

Capability 93 reflects the independently measured 45 versus 40 comparison and official claims of stronger coding and multimodal work. Autonomy 96 reflects documented long-horizon agent functionality but no independent evidence of unrestricted autonomy. Deployment 90 reflects listed international hosted access. Misuse 75 and control difficulty 80 inherit the conservative earlier exact-version baseline because this snapshot has no verified change on those dimensions. Availability remains unknown without affirmative version-specific artifact or proprietary-service proof.

Capability93
Autonomy96
Deployment90
Misuse potential75
Control difficulty80
MODEL-ATTRIBUTED EVIDENCE

News tied to Qwen3.8-Max-0902

The model score of 87.9 rates this model's risk profile. The overall Doom Index of 67.9 measures the complete temporally weighted evidence record. These values answer different questions.

NET MODEL-ATTRIBUTED INDEX CONTRIBUTION+0.07

Each article's current Doom Index contribution is divided equally among the exact models named on that article. This prevents multi-model evidence from being claimed in full on several model pages. Model risk scores use a bounded temporal offset around their technical profile, but never feed back into the overall index.

Autonomy TOWARD

Alibaba reports automated Qwen research cycles and a measured model gain

Alibaba says Qwen3.8-Max completed 33 automated AI research cycles over a month and a separate 60-hour chip-design run. It attributes a rise from 40 to 45 on Artificial Analysis to the research process; Artificial Analysis independently lists the September 2 Qwen3.8 Max snapshot at 45 versus 40 for the earlier version, but has not independently verified that causal account or the chip-design claims.

Full item contribution
+0.07
Qwen3.8-Max-0902 equal share
+0.07
Read assessment →
AUDIT TRAIL

Model score history

  1. R1
    Doom Score 87.9

    New official September 2 exact snapshot entry, independently compared at 45 versus 40 on the Artificial Analysis Intelligence Index.

    23 Sept 2026
SHARE THE FINDINGS

Share this page

DoomBench social sharing card for Qwen3.8-Max-0902.
  1. Qwen3.8-Max-0902 by Alibaba has a DoomBench model risk score of 87.9 out of 100, based on five transparent version-specific dimensions rather than the overall index.

  2. Qwen3.8-Max-0902's highest current DoomBench dimension is autonomy at 96.0 out of 100; the profile publishes every component score and its editorial rationale.

  3. DoomBench links 1 source-backed evidence item to Qwen3.8-Max-0902, while keeping the model's risk profile separate from each item's contribution to the live Doom Index.

    https://www.doombench.com/models/alibaba-qwen3-8-max-0902