Qwen3.8 27B Achieves Intelligence Index Score of 52, Ranks #1 Among 135 Models
- Published
- Aug 17, 2026 — 17:25 UTC
Qwen3.8 27B, released by Alibaba in August 2026, achieved an Intelligence Index score of 52, ranking #1 out of 135 models evaluated. The model has 27 billion parameters and features a context window size of 256k tokens. It generated 160 million output tokens at a cost of $0.00 per million input and output tokens, significantly undercutting the median costs of $0.04 and $0.15, respectively. This model was evaluated among 609 total models, with 9 evaluations contributing to its score. As noted by Hacker News (AI filtered), Qwen3.8 27B is positioned as a leading model in intelligence and competitively priced against other open-weight models of similar size.
By Callan Zhang · Aug 17, 2026 · Editorial standards →
Summarised from the primary source with AI assistance under human editorial oversight. Turing Wire is not a primary source — read the original for the authoritative account.
Source: Hacker News (AI filtered)