Alibaba
Qwen3.8 Flash
Release Date 2026-08-26(Knowledge cutoff 2026-03)1M IN · 131K OUTProprietary ModelAPI available
Qwen3.8 Flash is a multimodal reasoning model from Alibaba. It is suited for coding assistance, agentic workflows, visual understanding, document and codebase analysis, desktop interaction, chart analysis, and long-video analysis.
API cost
Provider price
$0.15IN$0.47OUT
SourceOpenRouterLast updated
Real-world usage
No.446 places0.3% share · 205B tokens/wk
295B9/7
460B9/14
509B9/21
505B9/28
205B10/5
Top use cases
- Math3.7%
SourceOpenRouterLast updated
Output Speed
Averageper 2,000 tokens
59tok/s
First Output 1.4stotal: 35.3s
SourceOpenRouterLast updated
User Comparisons
Writing Style
Answer length
Longer side
top 22%
Vocabulary diversity
Very diverse
top 19%
Sentence rhythm
Average
top 52%
Language mixing
Low
top 68%
List usage
Often
32% of lines
Heading usage
Occasionally
12% of lines
Reasoning share
Very heavy
68% of total
Response Speed
instability ±98%
First response 32.9sTotal 64.4s
44 tok/s
19.7ms per token
SourceAIB Inc.Last updated
Expert Performance Evaluation
Automation (agents)
All automation AI rankingsCan it handle customer service by the rules?τ³-BankingTop 11%#8
45.4%
Documents and office work
All documents and office work AI rankings79.7%74.255.7%16.6%
Image understanding
All image understanding and analysis AI rankingsCan it read charts and answer accurately?CharXivTop 10%#3
90.6%Self-reported by developer
Classification and judgment
All classification and judgment AI rankings62.462.4-1.37
See each evaluation’s detail page for its source and last updateData sources & removal requests
Related Info
Public
Alibaba🇨🇳 Hangzhou2023$267.3B~220,000 (total)