1 Pull request merged by 1 person

• updata faq file

  #744 merged Aug 7, 2025

6 Issues closed by 4 people

• Chunk too big

  #756 closed Aug 11, 2025

• 框架支持judge_strategy 扩展吗

  #760 closed Aug 7, 2025

• evalscope eval 命令行传入参数进行多模态模型评测数据集，出现错误

  #758 closed Aug 7, 2025

• 自定义数据集的时候assistant prompt在哪里可以添加

  #722 closed Aug 7, 2025

• MTEB两阶段评估是否支持API模式

  #727 closed Aug 7, 2025

• tau-bench评估api-key 报错

  #729 closed Aug 7, 2025

13 Issues opened by 11 people

• evalscope0.17.1新prompt导致部分模型humaneval精度review过低

  #768 opened Aug 12, 2025

• 使用 --use_cache 时新增指标未被计算

  #767 opened Aug 11, 2025

• 在evalscope如何中快速自定义RAGAS其他测评指标，想要自定义更多指标

  #766 opened Aug 11, 2025

• 文档中对--number参数的定义存在歧义

  #765 opened Aug 11, 2025

• multilingual_e5 embeddings模型压测失败！！！

  #764 opened Aug 8, 2025

• gsm8k的评测结果异常

  #763 opened Aug 8, 2025

• bfcl_v3测试评分基本都是0，Invalid syntax. Failed to decode AST.

  #762 opened Aug 7, 2025

• 多模态模型评测结果不对

  #761 opened Aug 7, 2025

• 无法看到评测进度，然后报cuda

  #759 opened Aug 7, 2025

• unstructured处理文档总是报错文档太短不能处理

  #757 opened Aug 6, 2025

• 要额外安装一堆依赖，wiki文档里为啥不写呀？或者在pip install的时候为啥不自动同时安装呀？

  #755 opened Aug 6, 2025

• 使用自定义数据集进行 T2I 评测时报错

  #754 opened Aug 6, 2025

• mteb task和evalscope task精度差异

  #753 opened Aug 6, 2025

3 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• bfcl_v3 数据集prompt长度超出模型的可输入长度

  #683 commented on Aug 7, 2025 • 0 new comments

• 解析response报错

  #751 commented on Aug 7, 2025 • 0 new comments

• [Refactor] v1.0

  #739 commented on Aug 12, 2025 • 0 new comments