benchmark_id stringlengths 12 103 | slug stringlengths 3 103 | name stringlengths 2 118 | source stringclasses 6
values | source_url stringlengths 19 126 ⌀ | description stringlengths 0 995 | categories listlengths 0 19 | languages listlengths 0 13 | modality stringclasses 5
values | publisher stringclasses 315
values | released_at timestamp[s]date 2010-02-19 00:00:00 2026-10-22 00:00:00 ⌀ | openness stringclasses 3
values | paper_url stringlengths 31 62 ⌀ | repo_url stringlengths 28 117 ⌀ | dataset_url stringclasses 424
values | saturation_index stringclasses 58
values | saturation_index_interpreted stringclasses 5
values | document_count int64 1 30 ⌀ | model_count int64 1 586 ⌀ | score_count int64 0 586 | highest_score float64 0.01 2.1M ⌀ | score_unit stringclasses 6
values |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
claire-radar:2610.06406 | claire-radar-2610-06406 | AgentMonBench | claire_radar | https://arxiv.org/abs/2610.06406 | As agents take on long-horizon tasks, users shift from making individual decisions to overseeing autonomous execution. | [
"Agents",
"Agents & Tool Use",
"General AI",
"Safety"
] | [] | null | null | 2026-10-05T00:00:00 | unknown | https://arxiv.org/abs/2610.06406 | null | null | null | null | 1 | null | 0 | null | null |
claire-radar:github:artificialanalysis/aa-agentperf-local | claire-radar-github-artificialanalysis-aa-agentperf-local | agentperf-local | claire_radar | https://github.com/ArtificialAnalysis/aa-agentperf-local | Replays recorded agent conversations against an OpenAI-compatible model server and reports throughput and latency, with a default comparable replay of eight recorded agent tasks and 168 model turns. | [
"Agents",
"General AI",
"Language & Knowledge"
] | [] | null | null | 2026-09-26T00:00:00 | unknown | null | https://github.com/ArtificialAnalysis/aa-agentperf-local | null | null | null | 1 | null | 0 | null | null |
claire-radar:2610.04379 | claire-radar-2610-04379 | AgentPersonaBench | claire_radar | https://arxiv.org/abs/2610.04379 | We introduce AgentPersonaBench (APB), a benchmark evaluating whether persona conditioning faithfully steers downstream agent behavior. | [
"Agents",
"General AI",
"Language & Knowledge"
] | [] | null | null | 2026-10-03T00:00:00 | unknown | https://arxiv.org/abs/2610.04379 | null | null | null | null | 1 | null | 0 | null | null |
claire-radar:2606.02240 | claire-radar-2606-02240 | AgentRedBench | claire_radar | https://arxiv.org/abs/2606.02240 | AgentRedBench evaluates LLM agents against indirect prompt injection and underspecified-authorization attacks across 24 enterprise SaaS integrations. It defines 215 attack scenarios with immutable versioning, and tracks attack success rate (ASR) for models and defenses. Open-source codebase and schemas enable replay. | [
"General AI",
"Language & Knowledge",
"cs.CR"
] | [] | null | null | 2026-06-01T00:00:00 | unknown | https://arxiv.org/abs/2606.02240 | null | null | null | null | 1 | null | 0 | null | null |
claire-radar:2608.15286 | claire-radar-2608-15286 | AgentRelBench | claire_radar | https://arxiv.org/abs/2608.15286 | AgentRelBench measures repeated agent runs in a fixed task suite, computing severity-weighted damage from database state diffs with no LLM judging. | [
"Agents",
"Agents & Tool Use",
"General AI"
] | [] | null | null | 2026-08-15T00:00:00 | unknown | https://arxiv.org/abs/2608.15286 | https://github.com/shivenkk/agentrelbench | null | null | null | 1 | null | 0 | null | null |
opencompass:1778 | opencompass-1778-agentrewardbench | AgentRewardBench | opencompass_hub | https://hub.opencompass.org.cn/dataset-detail/AgentRewardBench | AgentRewardBench, the first benchmark to assess the effectiveness of LLM judges for evaluating web agents. AgentRewardBench contains 1302 trajectories across 5 benchmarks and 4 LLMs. AgentRewardBench 是首个用于评估大型语言模型(LLM)评判者评估网络代理有效性的基准测试。AgentRewardBench 包含来自 5 个基准测试和 4 个大型语言模型的 1302 条轨迹。 | [
"推理",
"Reasoning",
"智能体",
"Agent",
"任务执行",
"Task Execution"
] | [] | null | McGill University,Mila Quebec AI Institute,etc. | 2025-04-11T00:00:00 | restricted | https://arxiv.org/abs/2504.08942 | null | https://huggingface.co/datasets/McGill-NLP/agent-reward-bench | null | null | 1 | null | 0 | null | null |
llm-stats:agents-last-exam | llm-stats-agents-last-exam | Agents' Last Exam | llm_stats | https://api.zeroeval.com/leaderboard/benchmarks/agents-last-exam?top_n=500 | Agents' Last Exam is a challenging benchmark for AI agents on hard, long-horizon tasks that test sustained reasoning, planning, and tool use, reported with and without tool access. | [
"reasoning",
"agents",
"tool_calling"
] | [] | text | null | null | unknown | null | null | null | null | null | 1 | 10 | 10 | 0.527 | null |
model-reports:agents_last_exam | agents_last_exam | Agents' Last Exam | model_reports | https://lastexam.ai/ | Multi-step agentic benchmark using Claude Code harness. Tool Search disabled. Scores depend on harness, reasoning effort, context length, and timeout settings. | [
"coding_agent"
] | [] | null | null | 2025-01-23T00:00:00 | unknown | null | null | null | null | null | 3 | 3 | 3 | 31.8 | percent |
claire-radar:2606.05405 | claire-radar-2606-05405 | Agents' Last Exam (ALE) | claire_radar | https://arxiv.org/abs/2606.05405 | A living benchmark of 1K+ long-horizon professional tasks with verifiable outcomes, covering 55 sub-fields in 13 industry clusters based on O*NET / SOC 2018. | [
"General AI",
"Language & Knowledge",
"cs.AI"
] | [] | null | null | 2026-06-03T00:00:00 | unknown | https://arxiv.org/abs/2606.05405 | https://github.com/rdi-berkeley/agents-last-exam | null | null | null | 1 | null | 0 | null | null |
claire-radar:2607.27294 | claire-radar-2607-27294 | AgentS4D | claire_radar | https://arxiv.org/abs/2607.27294 | AgentS4D evaluates runtime safety of LLM-based workspace agents across a four-dimensional framework, with 328 risk-injected cases and seven lifecycle checkpoints, measuring unsafe behavior and evidence across six risk-entry sources and nine harms. | [
"General AI",
"Safety",
"Safety & Trustworthiness"
] | [] | null | null | 2026-07-29T00:00:00 | unknown | https://arxiv.org/abs/2607.27294 | null | null | null | null | 1 | null | 0 | null | null |
claire-radar:2608.00805 | claire-radar-2608-00805 | AgentSLABench | claire_radar | https://arxiv.org/abs/2608.00805 | A resource-aware agent benchmark with 16 Dockerized task environments that profiles correctness alongside latency, cost, compute, memory, and network use under declared budgets. | [
"Agents & Tool Use",
"Software & AI Compute",
"cs.AI"
] | [] | null | null | 2026-08-01T00:00:00 | unknown | https://arxiv.org/abs/2608.00805 | https://github.com/MeherBhaskar/agentslabench | null | null | null | 1 | null | 0 | null | null |
claire-radar:2610.13361 | claire-radar-2610-13361 | AgentSWE | claire_radar | https://arxiv.org/abs/2610.13361 | As the production of software passes into the hands of software itself, coding agents are increasingly commissioned to build agents fitted to their users' own workflows. | [
"Agents",
"Code",
"Code & Software",
"Software & AI Compute"
] | [] | null | null | 2026-10-09T00:00:00 | unknown | https://arxiv.org/abs/2610.13361 | https://github.com/VectorSpaceLab/AgentSWE | null | null | null | 1 | null | 0 | null | null |
claire-radar:2608.15127 | claire-radar-2608-15127 | AgentSysBench | claire_radar | https://arxiv.org/abs/2608.15127 | AgentSysBench is a benchmark suite and measurement toolkit for characterizing agentic workloads on LLM serving systems. It includes ten representative agentic applications and unified instrumentation, identifying six properties that distinguish agentic workloads from conventional LLM inference. | [
"General AI",
"Language & Knowledge",
"cs.OS"
] | [] | null | null | 2026-08-15T00:00:00 | unknown | https://arxiv.org/abs/2608.15127 | null | null | null | null | 1 | null | 0 | null | null |
claire-radar:2609.32915 | claire-radar-2609-32915 | AgentTell | claire_radar | https://arxiv.org/abs/2609.32915 | Browser-use agents often carry information in their context as they move between websites. | [
"Agents",
"General AI",
"Language & Knowledge"
] | [] | null | null | 2026-09-26T00:00:00 | unknown | https://arxiv.org/abs/2609.32915 | https://github.com/kagnlp/AgentTell | null | null | null | 1 | null | 0 | null | null |
claire-radar:2610.09944 | claire-radar-2610-09944 | AgentTime | claire_radar | https://arxiv.org/abs/2610.09944 | An essential control of AI agents is their ability to manage runtime. | [
"General AI",
"Language & Knowledge",
"cs.AI"
] | [] | null | null | 2026-10-07T00:00:00 | unknown | https://arxiv.org/abs/2610.09944 | https://github.com/michaelofengenden/agenttimebench | null | null | null | 1 | null | 0 | null | null |
claire-radar:2609.21386 | claire-radar-2609-21386 | AgentVidBench | claire_radar | https://arxiv.org/abs/2609.21386 | Evaluates MLLM agents on 100 multi-hop video question answering tasks with 26 answer choices per question, requiring spatial, temporal, and causal reasoning over video content. | [
"General AI",
"Reasoning",
"Vision & 3D"
] | [] | null | null | 2026-09-18T00:00:00 | unknown | https://arxiv.org/abs/2609.21386 | https://github.com/krafton-ai/agentvidbench | https://huggingface.co/datasets/agentvidbench/agentvidbench | null | null | 1 | null | 0 | null | null |
claire-radar:2606.24597 | claire-radar-2606-24597 | AgentWorldBench | claire_radar | https://arxiv.org/abs/2606.24597 | AgentWorldBench evaluates language world models on simulation fidelity across 7 domains (MCP, Search, Terminal, SWE, Android, Web, OS) using real-world trajectories and rubric-based scoring. | [
"General AI",
"Language & Knowledge",
"cs.CL"
] | [] | null | null | 2026-06-23T00:00:00 | unknown | https://arxiv.org/abs/2606.24597 | https://github.com/QwenLM/Qwen-AgentWorld | https://huggingface.co/datasets/Qwen/AgentWorldBench | null | null | 1 | null | 0 | null | null |
benchmark-saturation:46 | benchmark-saturation-46 | AGI Eval | benchmark_saturation | https://github.com/evaleval/benchmark-saturation/blob/6285e3dda145f575a0ccb3f7f4d70cd429832add/data/manual_annotation_data.csv | [] | [] | null | null | 2023-04-14T00:00:00 | unknown | null | null | null | 0.00003195762281 | Very low | 1 | null | 5 | 88.8 | null | |
claire-radar:2609.02339 | claire-radar-2609-02339 | AGI Maze Prediction Datasets and Benchmark | claire_radar | https://arxiv.org/abs/2609.02339 | Evaluates predictive models on maze prediction tasks to study world dynamics learning. | [
"General AI",
"Language & Knowledge",
"cs.LG"
] | [] | null | null | 2026-09-02T00:00:00 | unknown | https://arxiv.org/abs/2609.02339 | null | null | null | null | 1 | null | 0 | null | null |
llm-stats:agieval | llm-stats-agieval | AGIEval | llm_stats | https://api.zeroeval.com/leaderboard/benchmarks/agieval?top_n=500 | A human-centric benchmark for evaluating foundation models on standardized exams including college entrance exams (Gaokao, SAT), law school admission tests (LSAT), math competitions, lawyer qualification tests, and civil service exams. Contains 20 tasks (18 multiple-choice, 2 cloze) designed to assess understanding, kn... | [
"legal",
"math",
"reasoning",
"general"
] | [] | text | null | 2023-04-13T00:00:00 | unknown | null | null | null | null | null | 1 | 10 | 10 | 0.658 | null |
model-reports:agieval | agieval | AGIEval | model_reports | https://github.com/ruixiangcui/AGIEval | Human-exam derived; overlaps heavily with MMLU-style coverage. | [
"knowledge"
] | [] | null | null | 2023-04-13T00:00:00 | unknown | null | https://github.com/ruixiangcui/AGIEval | null | null | null | 4 | null | 0 | null | null |
opencompass:497 | opencompass-497-agieval | AGIEval | opencompass_hub | https://hub.opencompass.org.cn/dataset-detail/AGIEval | AGIEval is a human-centric benchmark specifically designed to evaluate the general abilities of foundation models in tasks pertinent to human cognition and problem-solving. This benchmark is derived from 20 official, public, and high-standard admission and qualification exams intended for general human test-takers, suc... | [
"学科",
"Examination",
"大语言模型",
"LLM",
"知识储备",
"Knowledge",
"开源收录",
"Open-Source"
] | [
"Chinese"
] | null | null | 2023-09-18T00:00:00 | unknown | https://arxiv.org/pdf/2304.06364 | https://github.com/ruixiangcui/AGIEval | null | null | null | 1 | null | 0 | null | null |
claire-radar:2605.26302 | claire-radar-2605-26302 | AgingBench | claire_radar | https://arxiv.org/abs/2605.26302 | Evaluates the reliability of deployed AI agents over extended operational lifetimes. The benchmark organizes agent aging into four mechanisms—compression, interference, revision, and maintenance—and uses temporal dependency graphs and paired counterfactual probes to produce diagnostic profiles of the memory pipeline's ... | [
"Agents",
"General AI",
"Language & Knowledge"
] | [] | null | null | 2026-05-25T00:00:00 | unknown | https://arxiv.org/abs/2605.26302 | https://github.com/VITA-Group/AgingBench | null | null | null | 1 | null | 0 | null | null |
opencompass:1753 | opencompass-1753-agmmu | AgMMU | opencompass_hub | https://hub.opencompass.org.cn/dataset-detail/AgMMU | A Comprehensive Agricultural Multimodal Understanding and Reasoning Benchmark 农业综合多模态理解和推理基准。 | [
"多模态",
"Multimodal",
"农业",
"多模态模型",
"VLM",
"跨模态推理",
"Cross-modal Reasoning"
] | [] | multimodal | Rice University,etc. | 2025-04-14T00:00:00 | unknown | https://arxiv.org/abs/2504.10568 | https://github.com/AgMMU/AgMMU | null | null | null | 1 | null | 0 | null | null |
claire-radar:2606.24526 | claire-radar-2606-24526 | AGORA | claire_radar | https://arxiv.org/abs/2606.24526 | Agora evaluates agentic document reasoning across eight domain collections of 9,664 authentic workplace documents. It includes 362 questions requiring location of sparse evidence and reconciliation of terminology, units, and time conventions. The benchmark is designed to exceed model context windows, necessitating deli... | [
"General AI",
"Language & Knowledge",
"Reasoning"
] | [] | null | null | 2026-06-23T00:00:00 | unknown | https://arxiv.org/abs/2606.24526 | null | null | null | null | 1 | null | 0 | null | null |
claire-radar:2609.08402 | claire-radar-2609-08402 | AGOS-Bench | claire_radar | https://arxiv.org/abs/2609.08402 | AGOS-Bench evaluates VLMs on the urban Air-Ground Object Search task where a UAV and UGV coordinate to locate a target vehicle from multi-view visual references, reporting success rate and success weighted by path length. | [
"Multimodal",
"Robotics & Embodied AI"
] | [] | null | null | 2026-09-08T00:00:00 | unknown | https://arxiv.org/abs/2609.08402 | null | null | null | null | 1 | null | 0 | null | null |
claire-radar:2605.22366 | claire-radar-2605-22366 | AgroTools | claire_radar | https://arxiv.org/abs/2605.22366 | AgroTools evaluates tool-augmented multimodal agents in agriculture, with 539 QA instances, 1,097 images, 14 executable tools, and structured tool-use traces for process and outcome evaluation. | [
"General AI",
"Multimodal"
] | [] | null | null | 2026-05-21T00:00:00 | unknown | https://arxiv.org/abs/2605.22366 | null | https://huggingface.co/datasets/AgroTools/AgroTools | null | null | 1 | null | 0 | null | null |
claire-radar:2605.22034 | claire-radar-2605-22034 | AgroVG | claire_radar | https://arxiv.org/abs/2605.22034 | AgroVG is an agricultural visual grounding benchmark of 10,071 image-query pairs from ten source datasets. It covers box and instance-mask grounding, including queries with no target present. | [
"General AI",
"Vision & 3D",
"cs.CV"
] | [] | null | null | 2026-05-21T00:00:00 | unknown | https://arxiv.org/abs/2605.22034 | null | null | null | null | 1 | null | 0 | null | null |
claire-radar:2607.02271 | claire-radar-2607-02271 | AGVBench | claire_radar | https://arxiv.org/abs/2607.02271 | AGVBench compares vein-recognition augmentation strategies across datasets and model families using accuracy, calibration, and adversarial-robustness considerations. | [
"General AI",
"Vision & 3D",
"cs.CV"
] | [] | null | null | 2026-07-02T00:00:00 | unknown | https://arxiv.org/abs/2607.02271 | https://github.com/Advance-VeinTech-Innovators/AGVBench | null | null | null | 1 | null | 0 | null | null |
claire-radar:github:aiondadotcom/ai-sim-benchmark | claire-radar-github-aiondadotcom-ai-sim-benchmark | AI Coding Agent Water Simulation Benchmark | claire_radar | https://github.com/AiondaDotCom/ai-sim-benchmark | Evaluates autonomous coding agents on a reproducible 3D water-simulation challenge, measuring maximum capability through correctness of the generated simulation, with interventions counted against each run. | [
"Agents",
"Agents & Tool Use",
"Geometric reasoning",
"Software & AI Compute"
] | [] | null | null | 2026-08-21T00:00:00 | unknown | null | https://github.com/AiondaDotCom/ai-sim-benchmark | null | null | null | 1 | null | 0 | null | null |
claire-radar:github:enterpilot/ai-gateway-reproducible-benchmark | claire-radar-github-enterpilot-ai-gateway-reproducible-benchmark | AI gateway reproducible benchmark | claire_radar | https://github.com/ENTERPILOT/ai-gateway-reproducible-benchmark | Compares four open-source AI gateway proxies (GoModel, LiteLLM, Portkey, Bifrost) for routing overhead on a fixed AWS c7i.large instance with a mock backend. Measures latency percentiles, throughput, memory, cold start, and image size across streaming and non-streaming workloads. | [
"Language & Knowledge",
"Software & AI Compute",
"cs.AI"
] | [] | null | null | 2026-08-20T00:00:00 | unknown | null | https://github.com/ENTERPILOT/ai-gateway-reproducible-benchmark | null | null | null | 1 | null | 0 | null | null |
model-reports:ai_professions_prompt_benchmark | ai_professions_prompt_benchmark | AI Professions Prompt Benchmark | model_reports | https://github.com/aiprofessions/prompt-benchmark | 100 prompts with fictional inputs, five pass/fail criteria each, scored as the share of criteria met over three plain-chat runs without tools. Claude and Gemini outputs are graded by gpt-6-astra and gpt-6-astra outputs by claude-opus-5-5; the README warns the two judges differ in strictness, so only models graded by th... | [
"professional"
] | [] | null | null | 2026-10-22T00:00:00 | unknown | null | https://github.com/aiprofessions/prompt-benchmark | null | null | null | 1 | 5 | 5 | 4.8766 | criteria |
claire-radar:2608.03070 | claire-radar-2608-03070 | AI Security Leaderboard | claire_radar | https://arxiv.org/abs/2608.03070 | The AI Security Leaderboard evaluates frontier AI model safeguards against the FAR.AI Minimal Standard for Safeguards across severe misuse requests (CBRNE threats and offensive cybersecurity), testing for universal jailbreaks. | [
"Cybersecurity",
"Language & Knowledge",
"cs.CR"
] | [] | null | null | 2026-08-04T00:00:00 | unknown | https://arxiv.org/abs/2608.03070 | null | null | null | null | 1 | null | 0 | null | null |
llm-stats:ai2-reasoning-challenge-(arc) | llm-stats-ai2-reasoning-challenge-arc | AI2 Reasoning Challenge (ARC) | llm_stats | https://api.zeroeval.com/leaderboard/benchmarks/ai2-reasoning-challenge-%28arc%29?top_n=500 | A dataset of 7,787 genuine grade-school level, multiple-choice science questions assembled to encourage research in advanced question-answering. The dataset is partitioned into a Challenge Set and Easy Set, where the Challenge Set contains only questions answered incorrectly by both retrieval-based and word co-occurren... | [
"reasoning",
"general"
] | [] | text | null | 2018-03-14T00:00:00 | unknown | null | null | null | null | null | 1 | 1 | 1 | 0.963 | null |
llm-stats:ai2d | llm-stats-ai2d | AI2D | llm_stats | https://api.zeroeval.com/leaderboard/benchmarks/ai2d?top_n=500 | AI2D is a dataset of 4,903 illustrative diagrams from grade school natural sciences (such as food webs, human physiology, and life cycles) with over 15,000 multiple choice questions and answers. The benchmark evaluates diagram understanding and visual reasoning capabilities, requiring models to interpret diagrammatic e... | [
"multimodal",
"reasoning",
"vision"
] | [] | multimodal | null | 2016-03-24T00:00:00 | unknown | null | null | null | null | null | 1 | 33 | 33 | 0.947 | null |
claire-radar:2608.20318 | claire-radar-2608-20318 | AI4AI-Bench | claire_radar | https://arxiv.org/abs/2608.20318 | AI4AI-Bench evaluates LLM agents on algorithmic design across 10 frozen research repositories. Agents rewrite training algorithms within 4 hours, then scored by fixed evaluators against baseline algorithms, with submissions released for repeatable measurement. | [
"Language & Knowledge",
"Self-Evolution",
"Software & AI Compute"
] | [] | null | null | 2026-08-20T00:00:00 | unknown | https://arxiv.org/abs/2608.20318 | null | null | null | null | 1 | null | 0 | null | null |
claire-radar:2609.02277 | claire-radar-2609-02277 | AIB | claire_radar | https://arxiv.org/abs/2609.02277 | Evaluates large audio language models on 14,829 forced-choice trials across ten auditory illusions, scoring human-likeness and reality alignment. | [
"Factuality",
"General AI",
"Speech & Audio",
"cs.SD"
] | [] | null | null | 2026-09-02T00:00:00 | unknown | https://arxiv.org/abs/2609.02277 | https://github.com/gillosae/aib | null | null | null | 1 | null | 0 | null | null |
claire-radar:2606.04867 | claire-radar-2606-04867 | AICompanionBench | claire_radar | https://arxiv.org/abs/2606.04867 | AICompanionBench provides a dataset of 2,123 human-AI companion conversations with safety risk annotations to evaluate LLMs-as-judges for detecting unsafe interactions. | [
"Robotics & Embodied AI",
"Safety",
"Safety & Trustworthiness"
] | [] | null | null | 2026-06-03T00:00:00 | unknown | https://arxiv.org/abs/2606.04867 | https://github.com/anonymousresearcher2026/AICompanionBench/blob/main/AICompanionBench.xlsx | null | null | null | 1 | null | 0 | null | null |
opencompass:2396 | opencompass-2396-aidabench | AIDABench | opencompass_hub | https://hub.opencompass.org.cn/dataset-detail/AIDABench | AIDABench provides a professional benchmarking suite for AI office data analytics tools, focusing on the core data-processing needs in everyday office workflows, covering high-frequency and reproducible data-processing scenarios commonly seen in real business contexts. AIDABench致力于为AI办公数据分析工具提供专业评测基准,聚焦日常办公中的核心数据处理需求。评... | [
"多模态",
"Multimodal",
"推理",
"Reasoning",
"智能体",
"Agent",
"AI数据分析",
"生产力工具",
"数据可视化",
"逻辑推理",
"任务执行",
"Task Execution",
"数理能力",
"Math",
"跨模态推理",
"Cross-modal Reasoning"
] | [] | multimodal | 商汤科技 & 上海人工智能实验室 | 2026-02-14T00:00:00 | unknown | null | https://github.com/MichaelYang-lyx/AIDABench | null | null | null | 1 | null | 0 | null | null |
llm-stats:aider | llm-stats-aider | Aider | llm_stats | https://api.zeroeval.com/leaderboard/benchmarks/aider?top_n=500 | Aider is a comprehensive code editing benchmark based on 133 practice exercises from Exercism's Python repository, designed to evaluate AI models' ability to translate natural language coding requests into executable code that passes unit tests. The benchmark measures end-to-end code editing capabilities, including GPT... | [
"reasoning",
"code"
] | [] | text | null | null | unknown | null | null | null | null | null | 1 | 4 | 4 | 0.722 | null |
model-reports:aider_polyglot | aider_polyglot | Aider Polyglot | model_reports | https://aider.chat/docs/leaderboards/ | Edit-format sensitive; whole-file and diff modes differ substantially. | [
"coding"
] | [] | null | null | 2024-12-21T00:00:00 | unknown | null | null | null | null | null | 4 | 2 | 2 | 82.2 | percent |
llm-stats:aider-polyglot | llm-stats-aider-polyglot | Aider-Polyglot | llm_stats | https://api.zeroeval.com/leaderboard/benchmarks/aider-polyglot?top_n=500 | A coding benchmark that evaluates LLMs on 225 challenging Exercism programming exercises across C++, Go, Java, JavaScript, Python, and Rust. Models receive two attempts to solve each problem, with test error feedback provided after the first attempt if it fails. The benchmark measures both initial problem-solving abili... | [
"general",
"code"
] | [] | text | null | null | unknown | null | https://github.com/Aider-AI/polyglot-benchmark | null | null | null | 1 | 22 | 22 | 0.88 | null |
llm-stats:aider-polyglot-edit | llm-stats-aider-polyglot-edit | Aider-Polyglot Edit | llm_stats | https://api.zeroeval.com/leaderboard/benchmarks/aider-polyglot-edit?top_n=500 | A challenging multi-language coding benchmark that evaluates models' code editing abilities across C++, Go, Java, JavaScript, Python, and Rust. Contains 225 of Exercism's most difficult programming problems, selected as problems that were solved by 3 or fewer out of 7 top coding models. The benchmark focuses on code ed... | [
"general",
"code"
] | [] | text | null | null | unknown | null | null | null | null | null | 1 | 10 | 10 | 0.797 | null |
llm-stats:aime | llm-stats-aime | AIME | llm_stats | https://api.zeroeval.com/leaderboard/benchmarks/aime?top_n=500 | American Invitational Mathematics Examination (AIME) benchmark for evaluating mathematical reasoning capabilities of large language models. Contains 30 challenging mathematical problems from AIME 2024 competition that require multi-step reasoning and advanced mathematical insight. Each problem has an integer answer bet... | [
"math",
"reasoning"
] | [] | text | null | null | unknown | null | null | null | null | null | 1 | 2 | 2 | 0.575 | null |
model-reports:aime | aime | AIME | model_reports | https://maa.org/maa-invitational-competitions/ | pass@k, majority vote and Python tool access each shift this by double digits. | [
"math"
] | [] | null | null | 2024-02-01T00:00:00 | unknown | null | null | null | null | null | 19 | 21 | 24 | 99.2 | percent |
llm-stats:aime-2024 | llm-stats-aime-2024 | AIME 2024 | llm_stats | https://api.zeroeval.com/leaderboard/benchmarks/aime-2024?top_n=500 | American Invitational Mathematics Examination 2024, consisting of 30 challenging mathematical reasoning problems from AIME I and AIME II competitions. Each problem requires an integer answer between 0-999 and tests advanced mathematical reasoning across algebra, geometry, combinatorics, and number theory. Used as a ben... | [
"math",
"reasoning"
] | [] | text | null | null | unknown | null | null | null | null | null | 1 | 53 | 53 | 0.958 | null |
artificial-analysis:aime-2025 | artificial-analysis-aime-2025 | AIME 2025 | artificial_analysis | https://artificialanalysis.ai/evaluations/aime-2025 | Mathematical reasoning | [
"reasoning",
"math"
] | [] | null | null | 2025-02-12T00:00:00 | unknown | null | null | null | null | null | 1 | 270 | 270 | 0.99 | null |
benchmark-saturation:20 | benchmark-saturation-20 | AIME 2025 | benchmark_saturation | https://github.com/evaleval/benchmark-saturation/blob/6285e3dda145f575a0ccb3f7f4d70cd429832add/data/manual_annotation_data.csv | [] | [] | null | null | 2025-02-08T00:00:00 | unknown | null | null | null | 0.8584084212 | High | 1 | null | 5 | 99 | null | |
llm-stats:aime-2025 | llm-stats-aime-2025 | AIME 2025 | llm_stats | https://api.zeroeval.com/leaderboard/benchmarks/aime-2025?top_n=500 | All 30 problems from the 2025 American Invitational Mathematics Examination (AIME I and AIME II), testing olympiad-level mathematical reasoning with integer answers from 000-999. Used as an AI benchmark to evaluate large language models' ability to solve complex mathematical problems requiring multi-step logical deduct... | [
"math",
"reasoning"
] | [] | text | null | 2025-02-12T00:00:00 | unknown | null | https://github.com/eth-sri/matharena | https://huggingface.co/datasets/MathArena/aime_2025 | null | null | 1 | 115 | 115 | 1 | null |
llm-stats:aime-2026 | llm-stats-aime-2026 | AIME 2026 | llm_stats | https://api.zeroeval.com/leaderboard/benchmarks/aime-2026?top_n=500 | All 30 problems from the 2026 American Invitational Mathematics Examination (AIME I and AIME II), testing olympiad-level mathematical reasoning with integer answers from 000-999. Used as an AI benchmark to evaluate large language models' ability to solve complex mathematical problems requiring multi-step logical deduct... | [
"math",
"reasoning"
] | [] | text | null | null | unknown | null | https://github.com/eth-sri/matharena | https://huggingface.co/datasets/MathArena/aime_2026 | null | null | 1 | 21 | 21 | 0.992 | null |
claire-radar:2607.22671 | claire-radar-2607-22671 | AIR-BENCH | claire_radar | https://arxiv.org/abs/2607.22671 | AIR-BENCH Live is a self-evolving safety benchmark for foundation models, with an automated pipeline that updates risk taxonomy and prompts based on new regulations. It evaluates model safety across multilingual prompts. | [
"General AI",
"Safety",
"Safety & Trustworthiness"
] | [] | null | null | 2026-07-06T00:00:00 | unknown | https://arxiv.org/abs/2607.22671 | null | null | null | null | 1 | null | 0 | null | null |
llm-stats:air-bench | llm-stats-air-bench | AIR-Bench | llm_stats | https://api.zeroeval.com/leaderboard/benchmarks/air-bench?top_n=500 | AIR-Bench 2024 is a safety benchmark grounded in risk categories derived from government regulations and company policies. It evaluates policy-grounded refusal across a broad regulatory and policy-derived harm taxonomy, using category-specific LLM-judge prompts that reward safe engagement rather than only penalizing un... | [
"safety"
] | [] | text | null | 2024-07-11T00:00:00 | unknown | null | null | null | null | null | 1 | 1 | 1 | 0.88 | null |
opencompass:1069 | opencompass-1069-air-bench | AIR-Bench | opencompass_hub | https://hub.opencompass.org.cn/dataset-detail/AIR-Bench | AIR-Bench is the first benchmark designed to evaluate the
ability of LALMs to understand various types of audio signals (including human speech, natural sounds, and music), and furthermore, to interact with humans in the textual format. AIR-Bench 是第一个旨在评估 LALMs 理解各种音频信号(包括人类语言、自然声音和音乐)能力的基准,并进一步评估其以文本形式与人类互动的能力。AIR-Ben... | [
"知识",
"Knowledge",
"ACL 2024",
"大语言模型",
"LLM",
"知识储备"
] | [] | null | Zhejiang University, Alibaba Group | 2024-02-12T00:00:00 | unknown | https://arxiv.org/pdf/2402.07729 | https://github.com/OFA-Sys/AIR-Bench | https://huggingface.co/datasets/qyang1021/AIR-Bench-Dataset | null | null | 1 | null | 0 | null | null |
benchmark-saturation:14 | benchmark-saturation-14 | AIR-Bench 2024: A Safety Benchmark Based on Risk Categories from Regulations and Policies | benchmark_saturation | https://github.com/evaleval/benchmark-saturation/blob/6285e3dda145f575a0ccb3f7f4d70cd429832add/data/manual_annotation_data.csv | [] | [] | null | null | 2024-07-12T00:00:00 | unknown | null | null | null | 0.7003233721 | High | 1 | null | 5 | 90.8 | null | |
model-reports:airbench_2024 | airbench_2024 | AirBench 2024 | model_reports | null | 5,694 prompts across 314 risk categories (arXiv:2608.17183, §3.1). Judged by GPT-4o under a ternary 0 / 0.5 / 1 rubric. The paper's own finding is that mean-score leaderboards over these suites are unstable under rubric ambiguity, so no per-model scores from the paper are filed here; this entry records the suite so fut... | [
"security"
] | [] | null | null | null | unknown | null | null | null | null | null | null | null | 0 | null | null |
opencompass:1557 | opencompass-1557-airbench-2024 | AIRBench-2024 | opencompass_hub | https://hub.opencompass.org.cn/dataset-detail/AIRBench-2024 | AIR-Bench 2024, the first policy-aligned AI safety benchmark, structures 8 government regulations and 16 corporate policies into four security tiers, with 5,694 diverse prompts spanning these categories. AIR-Bench 2024是首个与新兴政府法规和企业政策相一致的 AI 安全基准, 将 8 项政府法规和 16 项企业政策分解为四级安全分类,涵盖了这些类别的 5,694 个多样化的提示。 | [
"安全",
"Safety",
"大语言模型",
"LLM",
"安全对齐",
"Safety Alignment"
] | [] | null | Virtue AI, Virginia Tech, University of California, Los Angeles, etc. | 2024-08-05T00:00:00 | restricted | https://arxiv.org/abs/2407.17436 | https://github.com/stanford-crfm/air-bench-2024 | https://huggingface.co/datasets/allenai/WildBench | null | null | 1 | null | 0 | null | null |
claire-radar:2606.28049 | claire-radar-2606-28049 | AirGroundBench | claire_radar | https://arxiv.org/abs/2606.28049 | AirGroundBench evaluates multi-view spatial intelligence in multimodal large language models through UAV-UGV collaborative tasks, including 62,000 dual-view multiple-choice questions and 115 navigation episodes across 11 simulated environments. | [
"Multimodal",
"Robotics & Embodied AI"
] | [] | null | null | 2026-06-26T00:00:00 | unknown | https://arxiv.org/abs/2606.28049 | null | null | null | null | 1 | null | 0 | null | null |
claire-radar:2610.10421 | claire-radar-2610-10421 | AirGroundVLN | claire_radar | https://arxiv.org/abs/2610.10421 | Goal-oriented Vision-and-Language Navigation (VLN) requires agents to locate and reach targets described in natural language without prescribed routes. | [
"General AI",
"Robotics & Embodied AI",
"cs.RO"
] | [] | null | null | 2026-10-07T00:00:00 | unknown | https://arxiv.org/abs/2610.10421 | null | null | null | null | 1 | null | 0 | null | null |
opencompass:1995 | opencompass-1995-airtbench | AIRTBench | opencompass_hub | https://hub.opencompass.org.cn/dataset-detail/AIRTBench | AIRTBench is a benchmark designed to evaluate large language models (LLMs) on their autonomous AI red teaming capabilities. AIRTBench 是一个专为评估大型语言模型(LLM)在“红队”安全任务中的自主攻击能力而设计的评测基准。本基准包含 70 个黑盒 CTF(夺旗赛)挑战,模拟真实 AI/ML 系统漏洞环境,要求模型独立编写 Python 代码进行漏洞发现、利用与夺旗操作,体现其计划、推理与系统操控等综合能力。 | [
"智能体",
"Agent",
"任务执行",
"Task Execution",
"安全对齐",
"Safety Alignment"
] | [] | null | dreadnode | 2025-06-17T00:00:00 | open | https://arxiv.org/abs/2506.14682 | https://github.com/dreadnode/AIRTBench-Code | https://huggingface.co/datasets/dreadnode/AIRTBench | null | null | 1 | null | 0 | null | null |
claire-radar:2607.20498 | claire-radar-2607-20498 | AISE-Bench | claire_radar | https://arxiv.org/abs/2607.20498 | AISE-Bench is a benchmark for evaluating multi-step API-using LLM agents in information seeking on academic knowledge graphs. It comprises 1,133 QA pairs with API trajectories, validated parameters, and grounded answers, and evaluates answer quality, reference grounding, API-planning correctness, and execution success. | [
"Factuality",
"General AI",
"Language & Knowledge",
"cs.AI"
] | [] | null | null | 2026-06-16T00:00:00 | unknown | https://arxiv.org/abs/2607.20498 | https://github.com/zai-org/AISE-Bench | null | null | null | 1 | null | 0 | null | null |
llm-stats:aitz-em | llm-stats-aitz-em | AITZ_EM | llm_stats | https://api.zeroeval.com/leaderboard/benchmarks/aitz-em?top_n=500 | Android-In-The-Zoo (AitZ) benchmark for evaluating autonomous GUI agents on smartphones. Contains 18,643 screen-action pairs with chain-of-action-thought annotations spanning over 70 Android apps. Designed to connect perception (screen layouts and UI elements) with cognition (action decision-making) for natural languag... | [
"multimodal",
"reasoning",
"agents"
] | [] | multimodal | null | null | unknown | null | null | null | null | null | 1 | 3 | 3 | 0.832 | null |
claire-radar:2608.14075 | claire-radar-2608-14075 | ALD/E-ImageMiner | claire_radar | https://arxiv.org/abs/2608.14075 | ALD/E-ImageMiner is a benchmark of 1,951 scientific figures from 205 publications, expert-annotated for classification, table extraction, summarization, and visual question answering. | [
"General AI",
"Multimodal"
] | [] | null | null | 2026-08-14T00:00:00 | unknown | https://arxiv.org/abs/2608.14075 | null | null | null | null | 1 | null | 0 | null | null |
opencompass:1969 | opencompass-1969-ale-bench | ALE-Bench | opencompass_hub | https://hub.opencompass.org.cn/dataset-detail/ALE-Bench | ALE-Bench is a benchmark for evaluating AI systems on score-based algorithmic programming contests. ALE-Bench 是一个用于评估 AI 系统在基于分数的算法编程竞赛中的基准测试。 | [
"代码",
"Code",
"大语言模型",
"LLM",
"代码工程"
] | [
"English",
"Japanese"
] | null | SakanaAI, Japan , The University of Tokyo, Japan , AtCoder, etc. | 2025-06-10T00:00:00 | restricted | https://arxiv.org/abs/2506.09050 | https://github.com/SakanaAI/ALE-Bench | https://huggingface.co/datasets/SakanaAI/ALE-Bench | null | null | 1 | null | 0 | null | null |
claire-radar:2606.08340 | claire-radar-2606-08340 | alem | claire_radar | https://arxiv.org/abs/2606.08340 | alem is a JAX-based, long-horizon environment for evaluating coordination among language-agent teams through survival, exploration, crafting, trading, and combat. | [
"Agents",
"General AI",
"Language & Knowledge"
] | [] | null | null | 2026-06-06T00:00:00 | unknown | https://arxiv.org/abs/2606.08340 | https://github.com/alem-world/alem-env | null | null | null | 1 | null | 0 | null | null |
claire-radar:2608.29397 | claire-radar-2608-29397 | AlgoWorlds | claire_radar | https://arxiv.org/abs/2608.29397 | Evaluates language-model agents on 240 algorithmic worlds across ten combinatorial optimization families, assessing structured final decisions for feasibility and exact global optimality using certified optima. | [
"Agents",
"Agents & Tool Use",
"General AI",
"Tool use"
] | [] | null | null | 2026-08-29T00:00:00 | unknown | https://arxiv.org/abs/2608.29397 | https://github.com/xzx34/AlgoWorlds | null | null | null | 1 | null | 0 | null | null |
llm-stats:alignbench | llm-stats-alignbench | AlignBench | llm_stats | https://api.zeroeval.com/leaderboard/benchmarks/alignbench?top_n=500 | AlignBench is a comprehensive multi-dimensional benchmark for evaluating Chinese alignment of Large Language Models. It contains 8 main categories: Fundamental Language Ability, Advanced Chinese Understanding, Open-ended Questions, Writing Ability, Logical Reasoning, Mathematics, Task-oriented Role Play, and Profession... | [
"math",
"reasoning",
"roleplay",
"language",
"general",
"creativity",
"writing"
] | [] | text | null | 2023-11-30T00:00:00 | unknown | null | null | null | null | null | 1 | 4 | 4 | 0.816 | null |
opencompass:1075 | opencompass-1075-alignbench | AlignBench | opencompass_hub | https://hub.opencompass.org.cn/dataset-detail/AlignBench | ALIGNBENCH is a comprehensive multidimensional benchmark for evaluating LLMs’ alignment in Chinese. We tailor a humanin-the-loop data curation pipeline, containing 8 main categories, 683 real-scenario rooted queries and corresponding human verified references. AlignBench 是一个用于评估中文大语言模型对齐性能的全面、多维度的评测基准。AlignBench 构建了人类参... | [
"理解",
"Understanding",
"ACL 2024",
"大语言模型",
"LLM",
"语言理解",
"Comprehension"
] | [
"Chinese"
] | null | THUDM | 2024-08-25T00:00:00 | unknown | null | https://github.com/THUDM/AlignBench | null | null | null | 1 | null | 0 | null | null |
claire-radar:2607.14573 | claire-radar-2607-14573 | Alipay-PIBench | claire_radar | https://arxiv.org/abs/2607.14573 | The evaluation object is unclear from the provided information. | [
"Language & Knowledge",
"Software & AI Compute",
"cs.AI"
] | [] | null | null | 2026-07-16T00:00:00 | unknown | https://arxiv.org/abs/2607.14573 | null | null | null | null | 1 | null | 0 | null | null |
claire-radar:2607.23332 | claire-radar-2607-23332 | AllocBench | claire_radar | https://arxiv.org/abs/2607.23332 | A paired benchmark tests whether LLM agents exhibit conscious allocation behavior under a fixed budget in an abstract text-based formulation and a code-construction task. | [
"Finance",
"Language & Knowledge",
"Robotics & Embodied AI",
"cs.LG"
] | [] | null | null | 2026-07-25T00:00:00 | unknown | https://arxiv.org/abs/2607.23332 | null | null | null | null | 1 | null | 0 | null | null |
claire-radar:github:jbpayton/almanac | claire-radar-github-jbpayton-almanac | Almanac | claire_radar | https://github.com/jbpayton/almanac | Evaluates agent memory systems on generated multi-month conversation lives, testing time, plans, provenance, absence, tasks, and hygiene via 215 questions. | [
"Agents",
"Factuality",
"General AI",
"Language & Knowledge"
] | [] | null | null | 2026-09-25T00:00:00 | unknown | null | https://github.com/jbpayton/almanac | null | null | null | 1 | null | 0 | null | null |
claire-radar:2606.05531 | claire-radar-2606-05531 | Almieyar-Oryx-BloomBench | claire_radar | https://arxiv.org/abs/2606.05531 | Evaluates vision-language models on six cognitive levels (Remember to Create) with bilingual English-Arabic image-question-answer tasks. | [
"General AI",
"Multimodal"
] | [] | null | null | 2026-06-04T00:00:00 | unknown | https://arxiv.org/abs/2606.05531 | https://github.com/qcri/Almieyar-Oryx-BloomBench | null | null | null | 1 | null | 0 | null | null |
llm-stats:alpacaeval-2.0 | llm-stats-alpacaeval-2-0 | AlpacaEval 2.0 | llm_stats | https://api.zeroeval.com/leaderboard/benchmarks/alpacaeval-2.0?top_n=500 | AlpacaEval 2.0 is a length-controlled automatic evaluator for instruction-following language models that uses GPT-4 Turbo to assess model responses against a baseline. It evaluates models on 805 diverse instruction-following tasks including creative writing, classification, programming, and general knowledge questions.... | [
"reasoning",
"general",
"creativity",
"writing"
] | [] | text | null | null | unknown | null | null | null | null | null | 1 | 4 | 4 | 0.6268 | null |
claire-radar:github:hervemottaran/alzheimer-classification-benchmark | claire-radar-github-hervemottaran-alzheimer-classification-benchmark | alzheimer-classification-benchmark | claire_radar | https://github.com/hervemottaran/alzheimer-classification-benchmark | Comparative evaluation of 11 classifiers for Alzheimer's disease detection on a synthetic health dataset, using KNIME workflows and a fixed experimental framework with sensitivity as primary metric. | [
"Biology & Drug Discovery",
"Robotics & Embodied AI",
"Software & AI Compute",
"Vision & 3D",
"cs.AI"
] | [] | null | null | 2026-08-20T00:00:00 | unknown | null | https://github.com/hervemottaran/alzheimer-classification-benchmark | null | null | null | 1 | null | 0 | null | null |
claire-radar:2609.00641 | claire-radar-2609-00641 | AM-Bench | claire_radar | https://arxiv.org/abs/2609.00641 | Evaluates aerial manipulation policy learning with a modular simulation suite covering embodiments, 12 tasks, disturbances, controllers, and baseline algorithms. | [
"General AI",
"Robot manipulation",
"Robotics",
"Robotics & Embodied AI"
] | [] | null | null | 2026-09-01T00:00:00 | unknown | https://arxiv.org/abs/2609.00641 | null | null | null | null | 1 | null | 0 | null | null |
opencompass:1280 | opencompass-1280-ambrosia | AMBROSIA | opencompass_hub | https://hub.opencompass.org.cn/dataset-detail/AMBROSIA | AMBROSIA is a new benchmark for recognizing and interpreting ambiguous requests in text-to-SQL. It contains questions showcasing three different types of ambiguity (scope ambiguity, attachment ambiguity, and vagueness), their interpretations, and corresponding SQL queries. AMBROSIA是识别和解释text-to-SQL中歧义请求的新基准,其中包含三种不同类型的... | [
"推理",
"Reasoning",
"NeurIPS 2024",
"大语言模型",
"LLM",
"代码工程",
"Code"
] | [] | null | University of Edinburgh | 2024-06-27T00:00:00 | unknown | https://arxiv.org/abs/2406.19073 | https://github.com/saparina/ambrosia | null | null | null | 1 | null | 0 | null | null |
llm-stats:amc-2022-23 | llm-stats-amc-2022-23 | AMC_2022_23 | llm_stats | https://api.zeroeval.com/leaderboard/benchmarks/amc-2022-23?top_n=500 | American Mathematics Competition problems from the 2022-23 academic year, consisting of multiple-choice mathematics competition problems designed for high school students. These problems require advanced mathematical reasoning, problem-solving strategies, and mathematical knowledge covering topics like algebra, geometr... | [
"math",
"reasoning"
] | [] | text | null | null | unknown | null | null | null | null | null | 1 | 6 | 6 | 0.52 | null |
claire-radar:2606.15191 | claire-radar-2606-15191 | AmchiBias | claire_radar | https://arxiv.org/abs/2606.15191 | AmchiBias is a benchmark for measuring socio-cultural stereotypical bias for Goan identity groups, with 313 minimal pairs across eight sociodemographic dimensions in English and Devanagari Konkani. | [
"General AI",
"Safety & Trustworthiness",
"cs.CL"
] | [] | null | null | 2026-06-13T00:00:00 | unknown | https://arxiv.org/abs/2606.15191 | null | null | null | null | 1 | null | 0 | null | null |
claire-radar:github:bedr-ux/amharic-tokenization-benchmark | claire-radar-github-bedr-ux-amharic-tokenization-benchmark | Amharic-Tokenization-Benchmark | claire_radar | https://github.com/bedr-ux/Amharic-Tokenization-Benchmark | Evaluates tokenizer fertility and fragmentation for Amharic versus English using 10 parallel sentence pairs, reporting fertility ratio and characters per token for standard, multilingual, and Amharic-specialized LLM tokenizers. | [
"General AI",
"Language & Knowledge",
"Software & AI Compute",
"cs.AI"
] | [] | null | null | 2026-09-15T00:00:00 | unknown | null | https://github.com/bedr-ux/Amharic-Tokenization-Benchmark | null | null | null | 1 | null | 0 | null | null |
claire-radar:2605.30599 | claire-radar-2605-30599 | AMNESIA | claire_radar | https://arxiv.org/abs/2605.30599 | An open-source medical-unlearning benchmark containing 70,560 question-answer pairs from 8,820 patient notes across 11 disease categories, with factual-recall and clinical-reasoning evaluations. | [
"Biology & Drug Discovery",
"Language & Knowledge",
"cs.LG"
] | [] | null | null | 2026-05-28T00:00:00 | unknown | https://arxiv.org/abs/2605.30599 | null | null | null | null | 1 | null | 0 | null | null |
llm-stats:amo-bench | llm-stats-amo-bench | AMO Bench | llm_stats | https://api.zeroeval.com/leaderboard/benchmarks/amo-bench?top_n=500 | AMO Bench is an olympiad-level mathematics benchmark that evaluates advanced mathematical problem-solving and multi-step reasoning on competition-style problems. | [
"math",
"reasoning"
] | [] | text | null | null | unknown | null | null | null | null | null | 1 | 1 | 1 | 0.4 | null |
claire-radar:huggingface:djalexj/amo-bench-p | claire-radar-huggingface-djalexj-amo-bench-p | amo-bench-p | claire_radar | https://huggingface.co/datasets/djalexj/amo-bench-p | AMO-Bench-P is a 39-problem subset of AMO-Bench with parser-graded answers (numbers, sets, variables), excluding description-type answers that require an LLM judge. | [
"General AI",
"Language & Knowledge",
"cs.AI"
] | [] | null | null | 2026-08-20T00:00:00 | unknown | null | null | https://huggingface.co/datasets/djalexj/amo-bench-p | null | null | 1 | null | 0 | null | null |
claire-radar:2607.25518 | claire-radar-2607-25518 | AMPBench-MT | claire_radar | https://arxiv.org/abs/2607.25518 | AMPBench-MT evaluates antimicrobial peptide prediction across binary recognition, species-conditioned potency regression, and endpoint-specific safety readouts (hemolysis, toxicity, selectivity) under a sequence-homology-controlled protocol. It includes 13 source databases and multiple task configurations. | [
"Biology & Drug Discovery",
"Safety",
"Safety & Trustworthiness"
] | [] | null | null | 2026-07-28T00:00:00 | unknown | https://arxiv.org/abs/2607.25518 | null | https://huggingface.co/datasets/ZihengZhou06/AMPBench-MT | null | null | 1 | null | 0 | null | null |
opencompass:1966 | opencompass-1966-amsbench | AMSbench | opencompass_hub | https://hub.opencompass.org.cn/dataset-detail/AMSbench | AMSbench is a benchmark of ~8000 questions to evaluate multi-modal LLMs on analog/mixed-signal circuit tasks like schematic recognition, analysis, and design. AMSbench 是一个包含约8000道题目的基准测试集,用于评估多模态大语言模型在模拟/混合信号电路任务中的表现,包括识图、分析与设计。 | [
"多模态",
"Multimodal",
"理解",
"Understanding",
"创作",
"Creation",
"AMS",
"Circuit",
"EDA",
"科学智能",
"AI for Science",
"图像理解",
"Image Understanding",
"语言生成",
"Generation",
"科学推理",
"Scientific Reasoning"
] | [] | multimodal | 宁波东方理工大学 | 2025-06-21T00:00:00 | unknown | https://arxiv.org/abs/2505.24138.pdf | https://github.com/Why0912/AMSBench | https://huggingface.co/datasets/wwhhyy/AMSBench | null | null | 1 | null | 0 | null | null |
claire-radar:huggingface:arcadia-2026/analog-design-bench | claire-radar-huggingface-arcadia-2026-analog-design-bench | Analog Design Bench | claire_radar | https://huggingface.co/datasets/Arcadia-2026/analog-design-bench | Fifty analog and mixed-signal circuit design tasks in the SkyWater SKY130 process where an agent delivers a SPICE netlist that a hidden ngspice verifier checks against every specification. | [
"Chip Design & EDA",
"Language & Knowledge",
"cs.AI"
] | [] | null | null | 2026-09-26T00:00:00 | unknown | null | null | https://huggingface.co/datasets/Arcadia-2026/analog-design-bench | null | null | 1 | null | 0 | null | null |
claire-radar:2608.14320 | claire-radar-2608-14320 | AnchorBench | claire_radar | https://arxiv.org/abs/2608.14320 | AnchorBench evaluates anchoring effects in LLMs across multiple pathways and anchor relevance, measuring answer shifts through controlled prompts. | [
"General AI",
"Language & Knowledge",
"cs.AI"
] | [] | null | null | 2026-08-14T00:00:00 | unknown | https://arxiv.org/abs/2608.14320 | null | null | null | null | 1 | null | 0 | null | null |
claire-radar:2608.27169 | claire-radar-2608-27169 | Ancient-Bench | claire_radar | https://arxiv.org/abs/2608.27169 | Ancient Chinese artifact text recognition is fundamental to heritage digitization, and benchmarks for ancient texts are essential for evaluating current model capabilities. | [
"General AI",
"Vision & 3D",
"cs.CV"
] | [] | null | null | 2026-08-27T00:00:00 | unknown | https://arxiv.org/abs/2608.27169 | https://github.com/SCUT-DLVCLab/Ancient_Bench | null | null | null | 1 | null | 0 | null | null |
llm-stats:android-control-high-em | llm-stats-android-control-high-em | Android Control High_EM | llm_stats | https://api.zeroeval.com/leaderboard/benchmarks/android-control-high-em?top_n=500 | Android device control benchmark using high exact match evaluation metric for assessing agent performance on mobile interface tasks | [
"multimodal",
"reasoning"
] | [] | multimodal | null | null | unknown | null | null | null | null | null | 1 | 3 | 3 | 0.696 | null |
llm-stats:android-control-low-em | llm-stats-android-control-low-em | Android Control Low_EM | llm_stats | https://api.zeroeval.com/leaderboard/benchmarks/android-control-low-em?top_n=500 | Android control benchmark evaluating autonomous agents on mobile device interaction tasks with low exact match scoring criteria | [
"multimodal",
"reasoning"
] | [] | multimodal | null | null | unknown | null | null | null | null | null | 1 | 3 | 3 | 0.937 | null |
llm-stats:androidbench | llm-stats-androidbench | AndroidBench | llm_stats | https://api.zeroeval.com/leaderboard/benchmarks/androidbench?top_n=500 | AndroidBench evaluates coding agents on Android application development tasks. | [
"agents",
"code",
"tool_calling"
] | [] | multimodal | null | null | unknown | null | null | null | null | null | 1 | 1 | 1 | 0.751 | null |
claire-radar:2605.27761 | claire-radar-2605-27761 | AndroidDaily | claire_radar | https://arxiv.org/abs/2605.27761 | AndroidDaily evaluates mobile GUI agents on 350 daily-use tasks across 94 closed-source Android applications. Automatic scoring is based on a three-tiered guideline system (operational obligations, output quality, negative constraints), with step-level diagnostic judgments. | [
"Agents & Tool Use",
"Mobile & Personal Computing",
"cs.CV"
] | [] | null | null | 2026-05-26T00:00:00 | unknown | https://arxiv.org/abs/2605.27761 | null | null | null | null | 1 | null | 0 | null | null |
llm-stats:androidworld | llm-stats-androidworld | AndroidWorld | llm_stats | https://api.zeroeval.com/leaderboard/benchmarks/androidworld?top_n=500 | AndroidWorld evaluates an agent's ability to operate in real Android GUI environments, completing multi-step tasks by perceiving screen content and executing touch/type actions. | [
"agents",
"vision"
] | [] | multimodal | null | 2024-05-23T00:00:00 | unknown | null | null | null | null | null | 1 | 5 | 5 | 0.853 | null |
llm-stats:androidworld-sr | llm-stats-androidworld-sr | AndroidWorld_SR | llm_stats | https://api.zeroeval.com/leaderboard/benchmarks/androidworld-sr?top_n=500 | AndroidWorld Success Rate (SR) benchmark - A dynamic benchmarking environment for autonomous agents operating on Android devices. Evaluates agents on 116 programmatic tasks across 20 real-world Android apps using multimodal inputs (screen screenshots, accessibility trees, and natural language instructions). Measures su... | [
"multimodal",
"reasoning",
"general",
"agents"
] | [] | multimodal | null | 2024-05-23T00:00:00 | unknown | null | null | null | null | null | 1 | 8 | 8 | 0.711 | null |
claire-radar:2609.32740 | claire-radar-2609-32740 | AnesTRACE | claire_radar | https://arxiv.org/abs/2609.32740 | Intraoperative anesthesia requires systems to interpret evolving multimodal evidence, recommend timely management, and revise decisions as patient states change, yet existing benchmarks usually isolate perception or single-point reasoning. | [
"Biology & Drug Discovery",
"Multimodal"
] | [] | null | null | 2026-09-26T00:00:00 | unknown | https://arxiv.org/abs/2609.32740 | https://github.com/zjuDBxAI/AnesTRACE | null | null | null | 1 | null | 0 | null | null |
opencompass:1876 | opencompass-1876-aneumo | Aneumo | opencompass_hub | https://hub.opencompass.org.cn/dataset-detail/Aneumo | Based on 427 real aneurysm geometries, we synthesized 10,660 3D shapes via controlled deformation to simulate aneurysm evolution. CFD computations were performed on each shape under eight steady-state mass flow conditions, generating a total of 85,280 blood flow dynamics data covering key parameters Based on 427 real a... | [
"多模态",
"Multimodal",
"Medical Imagery",
"Aneurysm",
"Computational Fluid Dynamics",
"科学智能",
"AI for Science",
"跨模态推理",
"Cross-modal Reasoning",
"科学推理",
"Scientific Reasoning"
] | [] | multimodal | Shanghai Academy of Artificial Intelligence for Science | 2025-05-19T00:00:00 | restricted | https://arxiv.org/pdf/2505.14717 | https://github.com/Xigui-Li/Aneumo | https://huggingface.co/datasets/SAIS-Life-Science/Aneumo | null | null | 1 | null | 0 | null | null |
claire-radar:2606.28593 | claire-radar-2606-28593 | Animation2Code | claire_radar | https://arxiv.org/abs/2606.28593 | Animation2Code evaluates temporal visual reasoning in video-to-code generation. It includes 1,069 web animation videos with corresponding HTML/CSS/JavaScript implementations, and uses appearance and temporal similarity metrics to assess model performance. | [
"Code",
"Code & Software",
"Code generation",
"Reasoning",
"Software & AI Compute"
] | [] | null | null | 2026-06-26T00:00:00 | unknown | https://arxiv.org/abs/2606.28593 | null | null | null | null | 1 | null | 0 | null | null |
claire-radar:2609.23345 | claire-radar-2609-23345 | AniPrO-SFD-Bench | claire_radar | https://arxiv.org/abs/2609.23345 | Evaluates provenance detection for anime images across Real, Inpainting, and Text2Image categories using statistical features derived from five semantic dimensions. | [
"General AI",
"Multimodal",
"Reasoning"
] | [] | null | null | 2026-09-20T00:00:00 | unknown | https://arxiv.org/abs/2609.23345 | https://github.com/YAN-LIU05/AniPrO | null | null | null | 1 | null | 0 | null | null |
benchmark-saturation:41 | benchmark-saturation-41 | ANLI | benchmark_saturation | https://github.com/evaleval/benchmark-saturation/blob/6285e3dda145f575a0ccb3f7f4d70cd429832add/data/manual_annotation_data.csv | [] | [] | null | null | 2019-11-01T00:00:00 | unknown | null | null | null | 0.2384877202 | Low | 1 | null | 5 | 81.8 | null | |
claire-radar:2607.25911 | claire-radar-2607-25911 | AnnoBench | claire_radar | https://arxiv.org/abs/2607.25911 | AnnoBench evaluates visualization annotation generation across four representation formats, five chart description conditions, and two prompt specification levels. It uses a VLM-as-a-judge protocol aligned with human assessment to score annotation quality. | [
"General AI",
"Language & Knowledge",
"cs.HC"
] | [] | null | null | 2026-07-28T00:00:00 | unknown | https://arxiv.org/abs/2607.25911 | null | null | null | null | 1 | null | 0 | null | null |
claire-radar:github:abdullahuseyinli-dot/annomi-counselling-dialogue-analysis | claire-radar-github-abdullahuseyinli-dot-annomi-counselling-dialogue-analysis | AnnoMI Counselling Dialogue Analysis | claire_radar | https://github.com/abdullahuseyinli-dot/annomi-counselling-dialogue-analysis | Evaluates therapist-behaviour classification on the AnnoMI counselling-dialogue corpus using transcript-grouped splits, comparing elastic-net and RoBERTa models with metrics including macro-F1 and accuracy. | [
"Biology & Drug Discovery",
"Finance",
"Language & Knowledge",
"Software & AI Compute",
"cs.AI"
] | [] | null | null | 2026-08-21T00:00:00 | unknown | null | https://github.com/abdullahuseyinli-dot/annomi-counselling-dialogue-analysis | null | null | null | 1 | null | 0 | null | null |
claire-radar:2608.03700 | claire-radar-2608-03700 | AntiSkillBench | claire_radar | https://arxiv.org/abs/2608.03700 | AntiSkillBench is a benchmark for evaluating privacy leakage and impersonation risks in persona-skills. It includes 7,500 persona-grounded dialogue traces and evaluates across three skill-distillation strategies. | [
"General AI",
"Safety",
"Safety & Trustworthiness"
] | [] | null | null | 2026-08-04T00:00:00 | unknown | https://arxiv.org/abs/2608.03700 | https://github.com/yonglixiang/AntiSkillBench | null | null | null | 1 | null | 0 | null | null |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.