DOOMBENCH.COM
Alibaba 🇨🇳 · Qwen2-VL

Qwen2-VL-72B-Instruct

The hosted 72-billion-parameter flagship Qwen2-VL model, released through an API with top-tier image, document, video, multilingual, and visual-agent capabilities.

DOOM SCORE64.7out 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.

Send detailed or private feedback →
CURRENT ASSESSMENT · REVISION 4

Why this model scores 64.7

Frontier-class visual reasoning and explicit agentic operation raise capability, autonomy, and misuse potential, while API-only availability moderates deployment but still leaves difficult multimodal control problems.

Capability70
Autonomy38
Deployment85
Misuse potential65
Control difficulty68
MODEL-ATTRIBUTED EVIDENCE

News tied to Qwen2-VL-72B-Instruct

The model score of 64.7 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+0.04

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 releases Qwen2-VL visual-agent models

Alibaba released open 2B and 7B Qwen2-VL models and a 72B API model with image, long-video, multilingual text, and visual-agent capabilities for operating mobile devices and robots from visual inputs.

Full item contribution
+0.12
Qwen2-VL-72B-Instruct equal share
+0.04
Read assessment →
AUDIT TRAIL

Model score history

  1. R4
    Doom Score 64.7

    Source-backed model availability audit using the model's existing primary or authoritative catalogue evidence. Exact-version evidence chronology replayed under temporal-monthly-pressure-v4.

    12 Aug 2026
  2. R3
    Doom Score 64.7

    Exact-version evidence chronology replayed after run doombench-hourly-news-20260812-142416 under temporal-monthly-pressure-v4.

    12 Aug 2026
  3. R2
    Doom Score 65.0

    Exact-version evidence chronology replayed after run doombench-hourly-news-20260812-132112 under fixed-sensitivity-v3.

    12 Aug 2026
  4. R1
    Doom Score 65.8

    New exact hosted Qwen2-VL flagship tier verified by the dated primary launch.

    12 Aug 2026
SHARE THE FINDINGS

Share this page

DoomBench social sharing card for Qwen2-VL-72B-Instruct.

  1. Qwen2-VL-72B-Instruct by Alibaba has a DoomBench model risk score of 64.7 out of 100, based on five transparent version-specific dimensions rather than the overall index.

  2. Qwen2-VL-72B-Instruct's highest current DoomBench dimension is deployment at 85.0 out of 100; the profile publishes every component score and its editorial rationale.

  3. DoomBench links 1 source-backed evidence item to Qwen2-VL-72B-Instruct, while keeping the model's risk profile separate from each item's contribution to the live Doom Index.

    https://www.doombench.com/models/alibaba-cloud-qwen2-vl-72b-instruct