{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:WEVYLKJ2LFQG5BOVBC6XITSRLT","short_pith_number":"pith:WEVYLKJ2","schema_version":"1.0","canonical_sha256":"b12b85a93a59606e85d508bd744e515cceb7b4606a32a9e3f14b4946475a0952","source":{"kind":"arxiv","id":"2207.00032","version":1},"attestation_state":"computed","paper":{"title":"DeepSpeed Inference: Enabling Efficient Inference of Transformer Models at Unprecedented Scale","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC","cs.PF"],"primary_cat":"cs.LG","authors_text":"Ammar Ahmad Awan, Cheng Li, Du Li, Elton Zheng, Jeff Rasley, Minjia Zhang, Olatunji Ruwase, Reza Yazdani Aminabadi, Samyam Rajbhandari, Shaden Smith, Yuxiong He","submitted_at":"2022-06-30T18:01:08Z","abstract_excerpt":"The past several years have witnessed the success of transformer-based models, and their scale and application scenarios continue to grow aggressively. The current landscape of transformer models is increasingly diverse: the model size varies drastically with the largest being of hundred-billion parameters; the model characteristics differ due to the sparsity introduced by the Mixture-of-Experts; the target application scenarios can be latency-critical or throughput-oriented; the deployment hardware could be single- or multi-GPU systems with different types of memory and storage, etc. With suc"},"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":"2207.00032","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-06-30T18:01:08Z","cross_cats_sorted":["cs.DC","cs.PF"],"title_canon_sha256":"c8a9437940f737f0058fa82f6352be1edd68bde16c02bac597af7e158300e037","abstract_canon_sha256":"b8231935d30144f2bb474f84003098b0853f940407010d836f0440e244359371"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:36:38.447291Z","signature_b64":"KDqB6iQcNL0mfnngvcv4jOjMfW/JtvFjSS7ymskYeY6uhvlM+a5B5ruo0RQIokkw4+pka8sOmtsmgV8A4AtCDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b12b85a93a59606e85d508bd744e515cceb7b4606a32a9e3f14b4946475a0952","last_reissued_at":"2026-07-05T04:36:38.446788Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:36:38.446788Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DeepSpeed Inference: Enabling Efficient Inference of Transformer Models at Unprecedented Scale","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC","cs.PF"],"primary_cat":"cs.LG","authors_text":"Ammar Ahmad Awan, Cheng Li, Du Li, Elton Zheng, Jeff Rasley, Minjia Zhang, Olatunji Ruwase, Reza Yazdani Aminabadi, Samyam Rajbhandari, Shaden Smith, Yuxiong He","submitted_at":"2022-06-30T18:01:08Z","abstract_excerpt":"The past several years have witnessed the success of transformer-based models, and their scale and application scenarios continue to grow aggressively. The current landscape of transformer models is increasingly diverse: the model size varies drastically with the largest being of hundred-billion parameters; the model characteristics differ due to the sparsity introduced by the Mixture-of-Experts; the target application scenarios can be latency-critical or throughput-oriented; the deployment hardware could be single- or multi-GPU systems with different types of memory and storage, etc. With suc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2207.00032","kind":"arxiv","version":1},"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/2207.00032/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":"2207.00032","created_at":"2026-07-05T04:36:38.446854+00:00"},{"alias_kind":"arxiv_version","alias_value":"2207.00032v1","created_at":"2026-07-05T04:36:38.446854+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2207.00032","created_at":"2026-07-05T04:36:38.446854+00:00"},{"alias_kind":"pith_short_12","alias_value":"WEVYLKJ2LFQG","created_at":"2026-07-05T04:36:38.446854+00:00"},{"alias_kind":"pith_short_16","alias_value":"WEVYLKJ2LFQG5BOV","created_at":"2026-07-05T04:36:38.446854+00:00"},{"alias_kind":"pith_short_8","alias_value":"WEVYLKJ2","created_at":"2026-07-05T04:36:38.446854+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":13,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.17566","citing_title":"AoiZora: Topology-Aware Auto-Parallel Optimization for Inference of Diffusion Transformers","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2607.01065","citing_title":"GSRQ: Gain-Shape Residual Quantization for Sub-1-bit KV Cache","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2606.27797","citing_title":"Optimizing Teacher-Student Partitioning for Scalable Knowledge Distillation on HPC Systems","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2605.27763","citing_title":"A Paired Testing Protocol for Batch-Conditioned Refusal Robustness in LLM Serving","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18750","citing_title":"A Readiness-Driven Runtime for Pipeline-Parallel Training under Runtime Variability","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2509.09505","citing_title":"Combating the Memory Walls: Optimization Pathways for Long-Context Agentic LLM Inference","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2306.14048","citing_title":"H$_2$O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Models","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2602.05695","citing_title":"SweetSpot: An Analytical Model for Predicting Energy Efficiency of LLM Inference","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04215","citing_title":"Predict-then-Diffuse: Adaptive Response Length for Compute-Budgeted Inference in Diffusion LLMs","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11111","citing_title":"ShardTensor: Domain Parallelism for Scientific Machine Learning","ref_index":57,"is_internal_anchor":false},{"citing_arxiv_id":"2309.06180","citing_title":"Efficient Memory Management for Large Language Model Serving with PagedAttention","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04215","citing_title":"Predict-then-Diffuse: Adaptive Response Length for Compute-Budgeted Inference in Diffusion LLMs","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2604.12358","citing_title":"Why and When Visual Token Pruning Fails? A Study on Relevant Visual Information Shift in MLLMs Decoding","ref_index":2,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WEVYLKJ2LFQG5BOVBC6XITSRLT","json":"https://pith.science/pith/WEVYLKJ2LFQG5BOVBC6XITSRLT.json","graph_json":"https://pith.science/api/pith-number/WEVYLKJ2LFQG5BOVBC6XITSRLT/graph.json","events_json":"https://pith.science/api/pith-number/WEVYLKJ2LFQG5BOVBC6XITSRLT/events.json","paper":"https://pith.science/paper/WEVYLKJ2"},"agent_actions":{"view_html":"https://pith.science/pith/WEVYLKJ2LFQG5BOVBC6XITSRLT","download_json":"https://pith.science/pith/WEVYLKJ2LFQG5BOVBC6XITSRLT.json","view_paper":"https://pith.science/paper/WEVYLKJ2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2207.00032&json=true","fetch_graph":"https://pith.science/api/pith-number/WEVYLKJ2LFQG5BOVBC6XITSRLT/graph.json","fetch_events":"https://pith.science/api/pith-number/WEVYLKJ2LFQG5BOVBC6XITSRLT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WEVYLKJ2LFQG5BOVBC6XITSRLT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WEVYLKJ2LFQG5BOVBC6XITSRLT/action/storage_attestation","attest_author":"https://pith.science/pith/WEVYLKJ2LFQG5BOVBC6XITSRLT/action/author_attestation","sign_citation":"https://pith.science/pith/WEVYLKJ2LFQG5BOVBC6XITSRLT/action/citation_signature","submit_replication":"https://pith.science/pith/WEVYLKJ2LFQG5BOVBC6XITSRLT/action/replication_record"}},"created_at":"2026-07-05T04:36:38.446854+00:00","updated_at":"2026-07-05T04:36:38.446854+00:00"}