Qwen3.8-27B open-weight model tops Hugging Face, scores 52 on AI Index

Alibaba's Qwen3.8-27B, an Apache 2 licensed 27B-parameter multimodal model, scores 52 on the Artificial Analysis Intelligence Index, matching GPT-5.6 Luna (max). It outperforms Qwen3.7-Plus and runs locally on consumer hardware, but defaults to 'xhigh' reasoning effort causing overthinking.
How this story unfolded
8 days · 6 reports · 29 community posts · 35 of 37 shown
- Aug 13
- Aug 14
- Aug 15
- Aug 16
- Aug 17
- Aug 18
- Aug 19
- Aug 20
- Aug 21
Qwen by email
Get an email when Qwen has news
No news that day, no email.
More stories today
- Reduce RAG costs on Amazon Bedrock with query-aware compression
- AWS and Panasonic Avionics use agentic AI for aircraft IFEC diagnostics
- User's Claude memory system backfired; Claude said user was the bottleneck
- Claude users discuss when to choose Sonnet over Opus
- Building token-efficient multi-agent systems