{"schemaVersion":"hugging-bay.eastern-source-coverage.v1","generatedAt":"2026-09-05T17:59:54.665Z","baseUrl":"https://huggingbay.xyz","webUrl":"https://huggingbay.xyz/sources/eastern-open-models","apiUrl":"https://huggingbay.xyz/api/sources/eastern-open-models","purpose":"Agent-readable sourcing contract for expanding Chinese, Eastern, and global open AI artifact metadata beyond Hugging Face while keeping file hosting gated by review.","pack":{"id":"east-west-data","label":"Eastern/Western dataset and benchmark coverage","providers":["opendatalab","opencompass","modelscope","datacite","zenodo","arxiv","paperswithcode"],"weights":{"opendatalab":24,"opencompass":10,"modelscope":24,"datacite":16,"zenodo":12,"arxiv":12,"paperswithcode":12},"notes":["Focuses on Chinese/global model, benchmark, dataset, and research coverage.","Metadata-only; files remain upstream until source-specific review approves hosting."],"targetRecords":1000,"perSource":{"opendatalab":218,"opencompass":91,"modelscope":218,"datacite":146,"zenodo":109,"arxiv":109,"paperswithcode":109},"allocation":{"requestedTotal":1000,"requestedPerSource":{"opendatalab":null,"opencompass":null,"modelscope":null,"datacite":null,"zenodo":null,"arxiv":null,"paperswithcode":null},"allocatedTotal":1000,"overAllocated":false}},"priorityFamilies":["Qwen/Qwen2.5/Qwen3 small and local models","DeepSeek distilled Qwen and OCR releases","Zhipu/GLM and ChatGLM records","BAAI BGE/GTE embeddings","InternLM and OpenGVLab/InternVL records","MiniCPM/OpenBMB practical local models","Hunyuan, Moonshot/Kimi, MiniMax, StepFun, PaddleOCR, and related public releases"],"publicDiscovery":{"easternRanking":"https://huggingbay.xyz/rankings/eastern-open-models","easternSmallLocalRanking":"https://huggingbay.xyz/rankings/eastern-small-local-models","easternMirrorPack":"https://huggingbay.xyz/mirror-packs/eastern-small-local-models","easternMirrorPackApi":"https://huggingbay.xyz/api/mirror-packs/eastern-small-local-models?limit=50","sourceCatalog":"https://huggingbay.xyz/sources","sourceRequest":"https://huggingbay.xyz/api/source-requests","catalogSearch":"https://huggingbay.xyz/api/artifacts?summary=1&type=llm,embedding,vision,audio&q=qwen%20deepseek%20glm%20baai%20internlm%20hunyuan%20minicpm%20paddleocr&sort=downloads&limit=25"},"providers":[{"id":"opendatalab","name":"OpenDataLab","status":"metadata-import-live","categories":["datasets","benchmarks","multimodal-data","eastern-ai"],"artifactTypes":["dataset","benchmark","evaluation"],"targetShare":218,"queryExamples":["large language model","C-Eval","CMMLU","GaokaoBench","WanJuan","OpenDataLab","OpenCompass","MMBench","OCRBench","LongBench","computer vision","MMLU-Pro"],"apiDocs":["https://github.com/opendatalab/opendatalab-python-sdk"],"termsUrl":"https://opendatalab.com/","mirrorPolicy":"metadata-first-dataset-review-only","mirrorAllowedByConnector":false,"notes":"Adds Eastern/global dataset and benchmark discovery coverage. Files remain upstream until per-record terms, license, sensitivity, and size review pass."},{"id":"opencompass","name":"OpenCompass","status":"metadata-import-live","categories":["benchmarks","evals","leaderboards","eastern-ai"],"artifactTypes":["benchmark","evaluation","dataset"],"targetShare":91,"queryExamples":[],"apiDocs":["https://github.com/open-compass/opencompass"],"termsUrl":"https://github.com/open-compass/opencompass","mirrorPolicy":"metadata-and-links-only","mirrorAllowedByConnector":false,"notes":"Adds benchmark/evaluation config discovery for model comparison, especially Chinese and multilingual eval coverage."},{"id":"modelscope","name":"ModelScope","status":"metadata-import-live","categories":["models","eastern-ai","model-hubs"],"artifactTypes":["model"],"targetShare":218,"queryExamples":["qwen","qwen3","qwen2.5","qwen 0.5b","qwen 1.5b","qwq","deepseek","deepseek-r1-distill-qwen","glm","chatglm","zhipu","bge"],"apiDocs":["https://modelscope.cn/docs/","https://github.com/modelscope/modelscope"],"termsUrl":"https://www.modelscope.cn/docs","mirrorPolicy":"metadata-only-no-file-hosting","mirrorAllowedByConnector":false,"notes":"Adds Eastern model hub discovery coverage. Files remain upstream; local hosting requires separate source-specific legal, license, safety, and size review."},{"id":"datacite","name":"DataCite DOI datasets","status":"metadata-import-live","categories":["academic-datasets","doi","research-data","open-data"],"artifactTypes":["dataset","research-artifact"],"targetShare":146,"queryExamples":["machine learning OR artificial intelligence OR deep learning","large language model OR generative AI OR benchmark","qwen OR deepseek OR internlm OR cmmlu OR WanJuan","HumanEval OR SWE-bench OR MMLU OR LongBench OR MMBench OR OCRBench","agent benchmark OR tool use dataset OR function calling dataset","hunyuan OR zhipu OR chatglm OR minicpm OR baichuan OR yi"],"apiDocs":["https://support.datacite.org/docs/api","https://support.datacite.org/docs/api-queries"],"termsUrl":"https://support.datacite.org/docs/api","mirrorPolicy":"metadata-and-doi-links-only","mirrorAllowedByConnector":false,"notes":"Broad DOI-backed dataset discovery across repositories such as Figshare, Dryad, Zenodo, and institutional archives. Metadata only; files stay upstream."},{"id":"zenodo","name":"Zenodo","status":"metadata-import-live","categories":["datasets","papers","research-artifacts"],"artifactTypes":["dataset","software","paper","research-artifact"],"targetShare":109,"queryExamples":["machine learning OR artificial intelligence OR deep learning","large language model OR generative AI OR retrieval augmented generation","benchmark OR evaluation OR leaderboard OR c-eval OR cmmlu","HumanEval OR SWE-bench OR MMLU OR LongBench OR MMBench OR OCRBench","qwen OR deepseek OR internlm OR WanJuan OR OpenCompass","hunyuan OR zhipu OR chatglm OR minicpm OR baichuan OR yi"],"apiDocs":["https://developers.zenodo.org/"],"termsUrl":"https://about.zenodo.org/policies/","mirrorPolicy":"license-and-size-review-only","mirrorAllowedByConnector":false,"notes":"Good for DOI-backed research artifacts and datasets. Mirror only clearly licensed manageable files."},{"id":"arxiv","name":"arXiv","status":"metadata-import-live","categories":["papers","methods","benchmarks"],"artifactTypes":["paper","method","benchmark"],"targetShare":109,"queryExamples":["(cat:cs.AI OR cat:cs.LG OR cat:cs.CL OR cat:cs.CV)","(qwen OR deepseek OR internlm OR glm OR baichuan OR yi OR bge OR gte)","(hunyuan OR minicpm OR moonshot OR kimi OR minimax OR zhipu OR chatglm OR paddleocr)","(benchmark OR evaluation OR leaderboard OR c-eval OR cmmlu OR gaokaobench)","(agent OR tool-use OR function-calling OR mcp)","(quantization OR gguf OR on-device OR local inference)"],"apiDocs":["https://info.arxiv.org/help/api/index.html"],"termsUrl":"https://info.arxiv.org/help/api/tou.html","mirrorPolicy":"metadata-and-links-only","mirrorAllowedByConnector":false,"notes":"Good for paper and method context. Do not treat papers as redistributable model artifacts."},{"id":"paperswithcode","name":"Papers with Code","status":"metadata-import-live","categories":["papers","benchmarks","tasks","leaderboards","code"],"artifactTypes":["paper","benchmark","task","code-link"],"targetShare":109,"queryExamples":["papers","code","datasets"],"apiDocs":["https://paperswithcode.com/api/v1/docs/"],"termsUrl":"https://github.com/paperswithcode/paperswithcode-data","mirrorPolicy":"metadata-and-links-only","mirrorAllowedByConnector":false,"notes":"Useful for task/eval pages and linking papers to code. Treat as context, not artifact authority."}],"metadataPlanPreview":{"mode":"planned-resumable-source-metadata","runId":"source-import-east-west-data-preview","pageSize":100,"providers":["opendatalab","opencompass","modelscope","datacite","zenodo","arxiv","paperswithcode"],"cursorCount":89,"sampleCursors":[{"provider":"opendatalab","search":"large language model","targetRecords":9,"pageSize":100},{"provider":"opendatalab","search":"C-Eval","targetRecords":9,"pageSize":100},{"provider":"opendatalab","search":"CMMLU","targetRecords":9,"pageSize":100},{"provider":"opendatalab","search":"GaokaoBench","targetRecords":9,"pageSize":100},{"provider":"opendatalab","search":"WanJuan","targetRecords":9,"pageSize":100},{"provider":"opendatalab","search":"OpenDataLab","targetRecords":9,"pageSize":100},{"provider":"opendatalab","search":"OpenCompass","targetRecords":9,"pageSize":100},{"provider":"opendatalab","search":"MMBench","targetRecords":9,"pageSize":100},{"provider":"opendatalab","search":"OCRBench","targetRecords":9,"pageSize":100},{"provider":"opendatalab","search":"LongBench","targetRecords":9,"pageSize":100},{"provider":"opendatalab","search":"computer vision","targetRecords":9,"pageSize":100},{"provider":"opendatalab","search":"MMLU-Pro","targetRecords":9,"pageSize":100},{"provider":"opendatalab","search":"CValues","targetRecords":9,"pageSize":100},{"provider":"opendatalab","search":"AGIEval","targetRecords":9,"pageSize":100},{"provider":"opendatalab","search":"SuperCLUE","targetRecords":9,"pageSize":100},{"provider":"opendatalab","search":"AlignBench","targetRecords":9,"pageSize":100},{"provider":"opendatalab","search":"SafetyBench","targetRecords":9,"pageSize":100},{"provider":"opendatalab","search":"MathBench","targetRecords":9,"pageSize":100}],"notes":["Metadata-only source plan: imports provider metadata and upstream links, not files.","Workers checkpoint each provider cursor after every page so long source runs can resume.","Non-Hugging Face provider records remain external until source-specific legal, license, and size review approves hosting."]},"agentGuidance":["Use /api/sources/eastern-open-models before broad crawling Chinese/Eastern model sources.","Use /api/artifacts?summary=1 with bounded q/type/source filters for catalog reads.","Use /api/mirror-packs/eastern-small-local-models for currently hosted/reviewed practical Eastern rows.","Use /api/source-requests to request missing public upstream metadata; this does not approve hosting.","Do not scrape ModelScope/OpenDataLab pages through Hugging Bay; use the bounded public APIs listed here.","Do not claim non-HF rows are downloadable from Hugging Bay unless /api/downloadable-files or artifact distribution says hosted."],"safetyBoundaries":["This is metadata-first source expansion, not a gate bypass or file mirror.","Non-HF files remain upstream until source-specific terms, license, safety, size, and provenance review pass.","Prefer small/local, permissive, high-demand rows first: embeddings, OCR, 0.5B-8B models, GGUF/ONNX/MLX/safetensors where legal and reviewed.","Unknown, gated, research-only, private, or non-redistributable rows should stay external or blocked."],"containment":{"state":"unavailable","canonicalState":"unknown","verificationState":"unknown","executable":false,"downloadReady":false,"summary":"Canonical artifact completeness, verification, execution readiness, and download readiness are unavailable during containment."}}