Unverified50% confidenceOpinionExact time

评测者认为通义千问3.6 27B是30B级别中对高考数学题表现最好的模型

1

Sources

50%

Confidence

Medium-term (~90 days)

Relevance

8/18/2026

First Seen

Valid until: 11/16/2026

Sources

Meta Muse Glimmer vs 通义千问：30B开源模型高考数学实测对比

Related Claims

Unverified本次评测使用全国一卷高考数学题对比Muse Glimmer和通义千问3.6 27B76% similar Unverified测试者评估Codex大概能完成看板上30%的任务70% similar Unverified通义千问3.6 27B在语义三次全对率和格式正确率上均领先Muse Glimmer70% similar UnverifiedQwen3.6-27B各项基准测试成绩在同规模（20B-30B参数级别）模型中处于领先地位68% similar Unverified测试者Luke认为Qwen3 27B是他测试过的在本地运行的最强编程模型63% similar

Cite This Claim

Stable URI

https://kongchang.com/claim/767869

API

curl https://kongchang.com/api/v1/knowledge/claims/767869

MCP

get_claim(id=767869)