{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:VCNH6EOGJPSKJBZ3CCEYISM2DB","short_pith_number":"pith:VCNH6EOG","schema_version":"1.0","canonical_sha256":"a89a7f11c64be4a4873b108984499a18748d5ccbfe6863002ac9864f1c25b6f3","source":{"kind":"arxiv","id":"2209.01188","version":2},"attestation_state":"computed","paper":{"title":"Petals: Collaborative Inference and Fine-tuning of Large Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.LG","authors_text":"Alexander Borzunov, Artem Chumachenko, Colin Raffel, Dmitry Baranchuk, Max Ryabinin, Pavel Samygin, Tim Dettmers, Younes Belkada","submitted_at":"2022-09-02T17:38:03Z","abstract_excerpt":"Many NLP tasks benefit from using large language models (LLMs) that often have more than 100 billion parameters. With the release of BLOOM-176B and OPT-175B, everyone can download pretrained models of this scale. Still, using these models requires high-end hardware unavailable to many researchers. In some cases, LLMs can be used more affordably via RAM offloading or hosted APIs. However, these techniques have innate limitations: offloading is too slow for interactive inference, while APIs are not flexible enough for research that requires access to weights, attention or logits. In this work, w"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2209.01188","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-09-02T17:38:03Z","cross_cats_sorted":["cs.DC"],"title_canon_sha256":"4d59fefc95f1539c1383fbfcb392c303a4f1597f0d0266c73623a809fca5dc0b","abstract_canon_sha256":"86c5f34bf32e2500aed2635eebee466aa26fb65f28e622ab8385b31ddb891e0a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:47:33.414300Z","signature_b64":"8ssr5q58SKj2LtJnkPXGYD5UBdFyAnKMqxIu0EcMl8QYd1KNIcORZaLdEr40+HdR7VrLYPwkkkrEyVlst/rbCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a89a7f11c64be4a4873b108984499a18748d5ccbfe6863002ac9864f1c25b6f3","last_reissued_at":"2026-07-05T05:47:33.413804Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:47:33.413804Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Petals: Collaborative Inference and Fine-tuning of Large Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.LG","authors_text":"Alexander Borzunov, Artem Chumachenko, Colin Raffel, Dmitry Baranchuk, Max Ryabinin, Pavel Samygin, Tim Dettmers, Younes Belkada","submitted_at":"2022-09-02T17:38:03Z","abstract_excerpt":"Many NLP tasks benefit from using large language models (LLMs) that often have more than 100 billion parameters. With the release of BLOOM-176B and OPT-175B, everyone can download pretrained models of this scale. Still, using these models requires high-end hardware unavailable to many researchers. In some cases, LLMs can be used more affordably via RAM offloading or hosted APIs. However, these techniques have innate limitations: offloading is too slow for interactive inference, while APIs are not flexible enough for research that requires access to weights, attention or logits. In this work, w"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2209.01188","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2209.01188/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2209.01188","created_at":"2026-07-05T05:47:33.413863+00:00"},{"alias_kind":"arxiv_version","alias_value":"2209.01188v2","created_at":"2026-07-05T05:47:33.413863+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2209.01188","created_at":"2026-07-05T05:47:33.413863+00:00"},{"alias_kind":"pith_short_12","alias_value":"VCNH6EOGJPSK","created_at":"2026-07-05T05:47:33.413863+00:00"},{"alias_kind":"pith_short_16","alias_value":"VCNH6EOGJPSKJBZ3","created_at":"2026-07-05T05:47:33.413863+00:00"},{"alias_kind":"pith_short_8","alias_value":"VCNH6EOG","created_at":"2026-07-05T05:47:33.413863+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.24722","citing_title":"Decentralised AI Training and Inference with BlockTrain","ref_index":75,"is_internal_anchor":false},{"citing_arxiv_id":"2605.28764","citing_title":"SwarmHarness: Skill-Based Task Routing via Decentralized Incentive-Aligned AI Agent Networks","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2604.21072","citing_title":"Distributed Generative Inference of LLM at Internet Scales with Multi-Dimensional Communication Optimization","ref_index":3,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VCNH6EOGJPSKJBZ3CCEYISM2DB","json":"https://pith.science/pith/VCNH6EOGJPSKJBZ3CCEYISM2DB.json","graph_json":"https://pith.science/api/pith-number/VCNH6EOGJPSKJBZ3CCEYISM2DB/graph.json","events_json":"https://pith.science/api/pith-number/VCNH6EOGJPSKJBZ3CCEYISM2DB/events.json","paper":"https://pith.science/paper/VCNH6EOG"},"agent_actions":{"view_html":"https://pith.science/pith/VCNH6EOGJPSKJBZ3CCEYISM2DB","download_json":"https://pith.science/pith/VCNH6EOGJPSKJBZ3CCEYISM2DB.json","view_paper":"https://pith.science/paper/VCNH6EOG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2209.01188&json=true","fetch_graph":"https://pith.science/api/pith-number/VCNH6EOGJPSKJBZ3CCEYISM2DB/graph.json","fetch_events":"https://pith.science/api/pith-number/VCNH6EOGJPSKJBZ3CCEYISM2DB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VCNH6EOGJPSKJBZ3CCEYISM2DB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VCNH6EOGJPSKJBZ3CCEYISM2DB/action/storage_attestation","attest_author":"https://pith.science/pith/VCNH6EOGJPSKJBZ3CCEYISM2DB/action/author_attestation","sign_citation":"https://pith.science/pith/VCNH6EOGJPSKJBZ3CCEYISM2DB/action/citation_signature","submit_replication":"https://pith.science/pith/VCNH6EOGJPSKJBZ3CCEYISM2DB/action/replication_record"}},"created_at":"2026-07-05T05:47:33.413863+00:00","updated_at":"2026-07-05T05:47:33.413863+00:00"}