According to CoinMeta, the qwen3.8-27b model officially open-sourced by Qianwen outperformed the newly released 30b model Meta in all 8 directly comparable tests listed by the official team. In the specific tests, qwen3.8-27b scored 73.0 in terminal-bench, Muse Glimmer scored 51.7, bench scored 61.7 against 51.2 in pro, and verified scored 84.3 against 65.9. In the comparison with Claude Opus which scored 4.6, qwen3.8-27b won 15 out of the tests, demonstrating its leadership in general reasoning, documentation, and visual tests.