HolyGrAIL

All tasks / Models

Qwen3.8-Max

Alibaba's hosted Qwen3.8-Max: the API version takes text, image and video input with a 1M-token context.

  • Alibaba (Qwen)
  • Closed
  • Released 2 Aug 2026

Where it ranks

  • A
    Chat & assistants
    Models · Cloud · 1,482 on LMArena Text Arena (overall, style control)
  • S
    Search & research
    Models · Cloud · 1,230 on LMArena Search Arena (style control)
  • A
    Writing
    Models · Cloud · 1,469 on LMArena Text Arena: Creative Writing (style control)
  • B
    Coding
    Models · Cloud · 1,674 on LMArena WebDev / Code Arena (overall)
  • S
    Image understanding
    Models · Cloud · 1,301 on LMArena Vision Arena (overall, style control)
  • C
    Workflow automation
    Models · Cloud · 56.2% on AutomationBench-AA (Artificial Analysis)

Facts

Vendor
Alibaba (Qwen)
Family
Qwen Max
Released
2 Aug 2026
Weights
Closed: API and apps only
License
proprietary
API price
$2 in / $6 out per 1M tokens (Alibaba Cloud Model Studio list price quoted by The Batch (cached input $0.25/1M); OpenRouter lists the same for the qwen3.8-max-0902 snapshot. The higher-throughput 'Max Prime' SKU (2026-09-23) costs $4/$12.)
Last verified
18 Oct 2026

Price from deeplearning.ai, 18 Oct 2026.

Benchmark results

BenchmarkResultEntry on the leaderboardAs ofSource
Artificial Analysis Intelligence Index45Qwen3.8 Max (0902)
Raw value from the page's embedded model data; the site displays it rounded to an integer.
19 Oct 2026artificialanalysis.ai
Artificial Analysis MMMU-Pro (independent evaluation)82.8%Qwen3.8 Max (0902)
Raw fraction 0.8277 from the model data embedded in https://artificialanalysis.ai/leaderboards/models, converted to percent; rank 19 of 124 entries with an MMMU-Pro result; no last-updated date shown.
19 Oct 2026artificialanalysis.ai
AutomationBench-AA (Artificial Analysis)56.2%Qwen3.8 Max (0902)
AutomationBench-AA Score (share of each task's objectives completed, a task with any guardrail violation scoring zero); raw value from the page's embedded model data (displayed rounded); rank 23 of 32; Tasks Completed 26.64%, Objectives Completed 87.84%; no last-updated date shown, so as_of = retrieval date.
19 Oct 2026artificialanalysis.ai
LMArena Search Arena (style control)1,230qwen3.8-max
From LMArena's official leaderboard-dataset (search_style_control, split latest, category overall, published 2026-10-13), which mirrors https://arena.ai/leaderboard/search (site blocks automated fetches). Rank 5 of 42; 1870 votes; 95% CI 1216.2-1242.8.
13 Oct 2026huggingface.co
LMArena Text Arena (overall, style control)1,482qwen3.8-max
From LMArena's official leaderboard-dataset (text_style_control, split latest, category overall, published 2026-10-15), which mirrors https://arena.ai/leaderboard/text (site blocks automated fetches). Rank 24 of 416; 27766 votes; 95% CI 1477.4-1487.3.
15 Oct 2026huggingface.co
LMArena Text Arena: Creative Writing (style control)1,469qwen3.8-max
From LMArena's official leaderboard-dataset (text_style_control, split latest, category creative_writing, published 2026-10-15), which mirrors https://arena.ai/leaderboard/text/creative-writing (site blocks automated fetches). Rank 19 of 414; 5471 votes; 95% CI 1460.0-1478.0.
15 Oct 2026huggingface.co
LMArena Vision Arena (overall, style control)1,301qwen3.8-max
From LMArena's official leaderboard-dataset (vision_style_control, split latest, category overall, published 2026-10-09), which mirrors https://arena.ai/leaderboard/vision (site blocks automated fetches). Rank 2 of 165; 10017 votes; 95% CI 1293.2-1308.1.
9 Oct 2026huggingface.co
LMArena WebDev / Code Arena (overall)1,674Qwen3.8-Max (0902)
From LMArena's official leaderboard-dataset (webdev = Code Arena, split latest, category overall, published 2026-10-18), which mirrors https://arena.ai/leaderboard/webdev. Rank 10; 14199 votes; 95% CI 1667.0-1681.1.
18 Oct 2026huggingface.co

Where you can use it

Compare

Alternatives to Qwen3.8-Max

Sources