Event ·Multi-source

Interfaze 开源 lev：基于 Qwen-3.5 4B 的判题模型

Interfaze 发布开源模型 lev，底座为 Qwen-3.5 4B。它能针对同一段文本一次性输出判断题、单选题和打分题的结果，不走逐字生成，而是直接读取模型内部概率。多道题目共享一次前向计算，耗时接近单次提问，且答案被限定在给定选项内。

Current read

Interfaze 开源了 lev，用 Qwen-3.5 4B 做底座，一次前向就能批改判断题、选择题和打分题，做自动测评的可以看看。

4 sources58° AI heatLast updated 2026/9/27 12:13:25

Evidence

4 sources

Primary sourceGorden Sun

2026/9/27 12:13:25View source

Related source 1AIGCLINK

2026/9/28 09:45:43View source

Related source 2OpenRouter

2026/9/25 19:02:06View source

Related source 3DAIR.AI

2026/9/27 22:35:02View source

Conflict check

Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.

Open the story