THE BENCHMARK DIRECTORY / 评测标准索引
AI 评测标准索引
任务定义、模型成绩，
以及它们背后的方法与来源。
精选评测
从任务到成绩，再到原始证据。
- Apodex 1.110.3%
- Claude Opus 517.5%
- DeepSeek V4 Flash-073112.4%
97 项任务 · Claude Code
查看全部 13 条成绩关于成绩预览
每张卡片预览三条来源顺序下的成绩，并非统一模型排名；横条保留各记录的原始单位与明确量程。GPQA Diamond 展示报告中的 pass@1 口径，FrontierChallenge 展示 Claude Code 配置。标签一致不代表条件完全相同，报告也可能引用其他机构的成绩。DeepSWE 1.1 未收录具体协议。进入详情可查看全部口径和逐条引用。日期表示采集或报告发布时间，不代表实时更新。
读懂结果，选对评测
全部文章 →浏览目录3,772 条记录
下载完整目录 ↓τ^τ-Bench: An Environment for End-To-End, Realistic Agent Construction
A benchmark where a developer agent builds a complete customer-service agent from business records, client requirements, a production API, inherited codebase, and cost/model limits, scored by deployment against held-out simulated users across 53 tasks in four domains.
2d-fet-bench
Field-effect transistor (FET) layouts on exfoliated two-dimensional flakes are typically drawn by hand for each flake, placing contacts and gates to match its position and outline in optical micrographs.
360CityArena
360CityArena evaluates embodied agents in a photorealistic virtual urban environment built from 360-degree video of Tokyo's Akihabara district. It includes 175 tasks across environment understanding, path reasoning, and spatial reasoning, testing localization, landmark search,…
3DCodeBench
3DCodeBench evaluates vision-language model agents on procedural 3D modeling by converting text and image references into Blender Python code. It includes 212 object categories with ground-truth scripts, and scores outputs on executability, image similarity (SigLIP-2/DINOv3),…
4DSynth-Nav
Evaluates embodied agents on interactive navigation tasks in procedurally generated 4D environments with independently tunable difficulty axes.
5-Dialects-BN
Evaluation object not clearly defined; abstract mentions alignment of Romanized transliteration with dialectal text, Standard Bangla, English, and subjectivity labels across five Bangla regional varieties, but no explicit task or scoring setup is provided.
A Large-Scale Dataset and Benchmark
InteractBind provides a large-scale dataset of ~100k protein-ligand pairs with fine-grained binding-site localization tasks and interaction maps for six non-covalent interaction types, plus affinity and similarity-controlled splits.
A Real-World CAD Segmentation Benchmark
CAD segmentation datasets are predominantly synthetic or automatically labelled, leaving it unclear how well existing methods transfer to real-world engineering tasks.
A-Bench
A-Bench是一个旨在诊断 LMMs 是否擅长评估 AIGIs 的基准，从 16 个文本到图像模型中采样了 2,864 个 AIGIs，每个都与由人类专家标注的问题-答案配对。
A-OKVQA
A-OKVQA用于评估多模态大模型的常识及推理能力，由25K个不同的问题组成，需要对图像中描述的场景进行某种形式的常识性推理来回答。
A3S-Bench
A3S-Bench is an executable security test suite for evaluating autonomous agents across multi-turn scenarios and agent-level risks using an action-grounded oracle.
AA-AnalystAgent
Quantitative analysis on spreadsheets & documents