OverviewAll LeaderboardsMethodologyComing soon
New Chat
Work IndexQwen3.8-27B
Alibaba

Qwen3.8-27B

Qwen3.8-27B is an open source model from Alibaba. It ranks #48 of 65 models on the Work Index with 1,024. Its best signal rank is #43 of 65, on Task Completion.

Work Index Overview

Rank
#48

Rank spread 44 – 53.

Work Index
1,024± 14
Pareto frontier

133 points under the frontier

Qwen3.8-27B
Developer
Alibaba
License
Apache 2.0
Token price (in / out)
$0.40 / $3 per 1M
Sessions
28,596
Ranking updated
Nov 13, 2026

Cost, Time, and Tokens

Cost per task$0.48
GPT-6 Luna (Low)$0.01AnthropicClaude Fable 5.1 (Max)$12.25

The median is $0.70.

Time per task8.2 min
MetaMuse Ember (Max)35 sAnthropicClaude Fable 5.2 (Max)25.1 min

The median is 8 min.

Output tokens per task73.4K
GPT-6 Luna (Low)5KAnthropicClaude Sonnet 5.5 (Max)174.1K

The median is 43.4K.

Work Ranking

9001,000
  1. 1,068
    GPT-5.6 TerraxHigh#41
  2. 1,065
    GPT-6 LunaMedium#42
  3. 1,061
    Stepfun
    Step 5 Preview#43
  4. 1,055
    MiniMax-M3.1-Flash#44
  5. 1,047
    GLM-5.3-Flash#45
  6. 1,037
    Qwen3.8-Flash-Next#46
  7. 1,035
    Gemini 3.7 FlashHigh#47
  8. 1,024
    Qwen3.8-27B#48
  9. 1,017
    Reflection AI
    Beam 1.0 501B A23B#49
  10. 1,013
    Gemini 3.1 Pro Preview#50
  11. 1,011
    GPT-6 LunaLow#51
  12. 1,003
    Mistral Large 4#52
  13. 1,002
    Tencent
    Hy3#53
  14. 982
    Qwen3.7-Max#54
  15. 970
    Qwen3.7-Plus#55

Signal Performance

Qwen3.8-27BLeader
Rank on the board
  • Task CompletionHow often the model gets users to confirm the task is done.
    Claude Sonnet 5.5 (Max) · 1,307Claude Sonnet 5.5 (Max) · 1,307Qwen3.8-27B · 1,076
    #43of 65
  • Praise vs ComplaintHow often the model earns more explicitly positive responses than negative ones.
    Claude Fable 5.2 · 1,861Claude Fable 5.2 · 1,861Qwen3.8-27B · 1,019
    #48of 65
  • SteerabilityHow well the model lands user corrections when they push back.
    Claude Fable 5.2 · 1,376Claude Fable 5.2 · 1,376Qwen3.8-27B · 1,003
    #49of 65
  • Bash ReliabilityHow often the model’s bash commands succeed. Errors outside its control aren’t counted. E.g. network issues, permission denied, etc.
    Claude Sonnet 5 · 1,234Claude Sonnet 5 · 1,234Qwen3.8-27B · 971
    #59of 65
9001,0001,1001,2001,3001,4001,5001,6001,7001,8001,900

USE CASES

  • Chat with AI
  • Build Apps & Websites
  • Write & Edit Text
  • Search the Web
  • Generate Images
  • Generate Videos
  • Chose any model
  • Compare Models Side by Side

LEADERBOARD RANKINGS

  • Overall
  • Agent
  • Alignment Index
  • Text
  • WebDev
  • Image-to-WebDev
  • Text to Image
  • Image Edit
  • Text to Video
  • Image to Video
  • Video Edit
  • Vision
  • Document
  • Search

COMPANY

  • About Us
  • How It Works
  • Blog
  • Careers
  • Leaderboard Changelog
  • Product Changelog
  • Help Center
  • FAQ

LEGAL

  • Terms
  • Privacy
  • Cookies

FOLLOW

  • X
  • LinkedIn
  • YouTube
  • Discord

© Arena Intelligence