| benchmark_id,benchmark_category |
| agi-eval,Reasoning and knowledge |
| agi-eval,Math |
| ai2-arc,Reasoning and knowledge |
| ai2d,Audio-visual understanding |
| aider-polyglot,Coding |
| aime-24,Math |
| aime-25,Math |
| aitz,Tool orchestration |
| alpaca-eval,Instruction following |
| amc,Math |
| android-control,Tool orchestration |
| android-world,Tool orchestration |
| arc-agi-2,Reasoning and knowledge |
| arc-agi-2,Audio-visual understanding |
| arc-challenge,Reasoning and knowledge |
| arena-hard-auto,Generic |
| asdiv,Math |
| basic-skills,Base model capabilities |
| bbh,Reasoning and knowledge |
| bbh,Math |
| bcb,Coding |
| bfcl,Tool orchestration |
| bfcl-3,Tool orchestration |
| bird-sql,Coding |
| blink,Audio-visual pattern recognition |
| browse-comp,Strategic planning |
| browse-comp-plus,Strategic planning |
| browse-comp-zh,Strategic planning |
| c-eval,Reasoning and knowledge |
| c-eval,Math |
| c-math,Math |
| c-simple-qa,Factuality |
| c3,Audio-visual understanding |
| cc-bench,Reasoning and knowledge |
| cc-bench,Audio-visual understanding |
| cc-ocr,Audio-visual pattern recognition |
| ccpm,Reasoning and knowledge |
| cfb,Rule adherence |
| cfb,Strategic planning |
| charades-sta,Audio-visual understanding |
| chart-qa,Audio-visual understanding |
| charvix,Audio-visual understanding |
| chinese-word,Multimodal generation |
| clasi,Translation |
| clasi,Audio-visual pattern recognition |
| clue-wsc,Commonsense |
| cm17k,Math |
| cmmlu,Reasoning and knowledge |
| cmmlu,Math |
| cmrc,Reasoning and knowledge |
| cmteb,Reasoning and knowledge |
| cmteb,Semantic search |
| cnmo,Math |
| codeforces,Coding |
| collie,Instruction following |
| coqa,Reasoning and knowledge |
| count-bench,Audio-visual pattern recognition |
| covost2,Translation |
| creative-writing,Writing style |
| crux-eval,Coding |
| cs-qa,Commonsense |
| cv-15,Audio-visual pattern recognition |
| cv-bench,Audio-visual pattern recognition |
| cv-bench,Audio-visual understanding |
| deepmind-math,Math |
| deepseek-leetcode,Coding |
| doc-vqa,Audio-visual pattern recognition |
| dpg,Multimodal generation |
| drop,Reasoning and knowledge |
| ds-1000,Coding |
| ego-schema,Audio-visual understanding |
| eq-bench,Reasoning and knowledge |
| erqa,Embodied spatial understanding |
| eval-plus,Coding |
| fact-score,Factuality |
| facts,Factuality |
| facts,Base model capabilities |
| finance-agent,Strategic planning |
| finance-agent,Tool orchestration |
| fleurs,Translation |
| fleurs,Audio-visual pattern recognition |
| follow-ir,Instruction following |
| frames,Factuality |
| frames,Reasoning and knowledge |
| frontier-math,Math |
| full-stack-bench,Coding |
| full-stack-bench,Math |
| gdp-val,Tool orchestration |
| gdp-val,Reasoning and knowledge |
| gedit-bench,Multimodal generation |
| gen-eval,Multimodal generation |
| global-mmlu,Reasoning and knowledge |
| global-mmlu,Multilingual performance |
| global-mmlu,Math |
| global-piqa,Commonsense |
| global-piqa,Multilingual performance |
| gpqa-diamond,Reasoning and knowledge |
| gpqa-diamond,Math |
| gpqa-main,Reasoning and knowledge |
| gpqa-main,Math |
| gpqa-science,Reasoning and knowledge |
| gpqa-science,Math |
| graph-walks,Long-context |
| graph-walks,Coding |
| gsm-symbolic,Math |
| gsm8k,Math |
| gso,Math |
| gtzan,Audio-visual pattern recognition |
| hallusion-bench,Audio-visual pattern recognition |
| hallusion-bench,Audio-visual understanding |
| health-bench-hard,Health |
| heatlh-bench,Health |
| hellaswag,Commonsense |
| hidden-math,Math |
| hle,Reasoning and knowledge |
| hle,Math |
| hle,Audio-visual understanding |
| hmmt,Math |
| hr-bench,Audio-visual pattern recognition |
| human-eval,Coding |
| human-eval-plus,Coding |
| if-bench,Instruction following |
| if-eval,Instruction following |
| img-edit,Multimodal generation |
| img-edit,Instruction following |
| imo-answer-bench,Math |
| include,Reasoning and knowledge |
| include,Multilingual performance |
| include,Math |
| info-vqa,Audio-visual understanding |
| jeopardy,Factuality |
| lambada,Base model capabilities |
| lbpp,Coding |
| lcb,Coding |
| lcb-pro,Coding |
| libri-speech,Audio-visual pattern recognition |
| lit-qa,Factuality |
| lit-qa,Long-context |
| live-bench,Reasoning and knowledge |
| live-bench,Coding |
| live-bench,Instruction following |
| loft,Long-context |
| loft,Audio-visual pattern recognition |
| long-align,Long-context |
| long-bench-2,Long-context |
| long-bench-2,Reasoning and knowledge |
| long-fact,Factuality |
| long-video-bench,Audio-visual understanding |
| ltb,Multimodal generation |
| lv-bench,Audio-visual understanding |
| lv-eval,Long-context |
| math,Math |
| math-500,Math |
| math-arena-apex,Math |
| math-instruct,Math |
| math-qa,Math |
| math-v,Math |
| math-v,Audio-visual understanding |
| math-verse,Math |
| math-verse,Audio-visual understanding |
| math-vista,Math |
| math-vista,Audio-visual understanding |
| mbpp,Coding |
| mbpp-plus,Coding |
| mcp-atlas,Tool orchestration |
| mcp-mark,Tool orchestration |
| mcp-mark,Strategic planning |
| mcp-universe,Tool orchestration |
| mcp-universe,Strategic planning |
| mcp-universe,Long-context |
| med-mc-qa,Reasoning and knowledge |
| med-qa,Reasoning and knowledge |
| mega-bench,Audio-visual understanding |
| mgsm,Math |
| mgsm,Multilingual performance |
| mini-f2f,Math |
| mini-max,Multimodal generation |
| mldr,Semantic search |
| mldr,Long-context |
| mlvu,Audio-visual understanding |
| mlvu,Long-context |
| mm-bench,Audio-visual understanding |
| mm-bench,Audio-visual pattern recognition |
| mm-bench-v,Audio-visual understanding |
| mm-bench-v,Audio-visual pattern recognition |
| mm-mt-bench,Instruction following |
| mm-mt-bench,Audio-visual understanding |
| mm-star,Audio-visual pattern recognition |
| mm-star,Audio-visual understanding |
| mmau,Audio-visual understanding |
| mmcu,Reasoning and knowledge |
| mmlu,Reasoning and knowledge |
| mmlu,Math |
| mmlu-pro,Reasoning and knowledge |
| mmlu-pro,Math |
| mmlu-pro-x,Reasoning and knowledge |
| mmlu-pro-x,Math |
| mmlu-pro-x,Multilingual performance |
| mmlu-redux,Reasoning and knowledge |
| mmlu-redux,Math |
| mmmlu,Reasoning and knowledge |
| mmmlu,Math |
| mmmlu,Multilingual performance |
| mmmu,Reasoning and knowledge |
| mmmu,Math |
| mmmu,Audio-visual understanding |
| mmmu-pro,Reasoning and knowledge |
| mmmu-pro,Math |
| mmmu-pro,Audio-visual understanding |
| mmteb,Semantic search |
| mmteb,Multilingual performance |
| mrcr,Long-context |
| mt-bench,Reasoning and knowledge |
| mteb,Semantic search |
| mteb,Multilingual performance |
| mtob,Translation |
| mu-cho-music,Audio-visual understanding |
| multi-challenge,Instruction retention |
| multi-challenge,Instruction following |
| multi-if,Instruction following |
| multi-pl-e,Coding |
| musr,Reasoning and knowledge |
| mv-bench,Audio-visual understanding |
| natural-questions,Reasoning and knowledge |
| ocr-bench,Audio-visual pattern recognition |
| ocr-bench,Audio-visual understanding |
| olympic-arena,Reasoning and knowledge |
| olympic-arena,Math |
| olympic-arena,Audio-visual understanding |
| omega,Math |
| omni,Audio-visual understanding |
| omni-doc-bench,Audio-visual understanding |
| one-ig-bench,Multimodal generation |
| openai-mrcr,Long-context |
| os-world,Tool orchestration |
| pile-test,Base model capabilities |
| piqa,Commonsense |
| pop-qa,Factuality |
| prover-bench,Math |
| putnam-bench,Math |
| race,Coding |
| real-world-qa,Embodied spatial understanding |
| research-eval,Strategic planning |
| ruler,Long-context |
| ruler,Reasoning and knowledge |
| sci-code,Coding |
| sci-code,Math |
| sci-q,Reasoning and knowledge |
| screen-spot,Audio-visual understanding |
| screen-spot-pro,Audio-visual understanding |
| seed-tts-eval,Multimodal generation |
| sifo,Instruction following |
| simple-qa,Factuality |
| simple-vqa,Factuality |
| simple-vqa,Audio-visual understanding |
| social-qa,Commonsense |
| squad,Reasoning and knowledge |
| super-gpqa,Reasoning and knowledge |
| super-gpqa,Math |
| svamp,Math |
| swe-bench-multilingual,Coding |
| swe-bench-pro,Coding |
| swe-bench-verified,Coding |
| swe-lancer-diamond,Coding |
| sysbench,Instruction retention |
| table-vqa,Audio-visual understanding |
| tau-bench,Rule adherence |
| tau2-bench,Rule adherence |
| temp-compass,Audio-visual understanding |
| terminal-bench,Tool orchestration |
| text-craft,Multimodal generation |
| tool-decathlon,Tool orchestration |
| trivia-qa,Reasoning and knowledge |
| truth-qa,Factuality |
| usamo-25,Math |
| vending-bench,Strategic planning |
| vending-bench-2,Strategic planning |
| vibe-eval,Audio-visual understanding |
| video-mme,Audio-visual understanding |
| video-mmmu,Reasoning and knowledge |
| video-mmmu,Audio-visual understanding |
| visu-logic,Audio-visual understanding |
| visual-probe,Audio-visual pattern recognition |
| visual-puzzles,Audio-visual understanding |
| vlms-blind,Audio-visual pattern recognition |
| voice-bench,Audio-visual understanding |
| vstar,Audio-visual understanding |
| wenet-speech,Audio-visual pattern recognition |
| wild-bench,Generic |
| winogrande,Commonsense |
| wmt-24,Translation |
| world-sense,Audio-visual understanding |
| world-sense,Audio-visual pattern recognition |
| writing-bench,Writing style |
| x-browse,Strategic planning |
| xbench-deepsearch,Strategic planning |
| zebra-logic,Reasoning and knowledge |
| zero-bench,Audio-visual understanding |
|
|
