{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:MUXVZ2YZPGB77M3O6OENS3H22Y","short_pith_number":"pith:MUXVZ2YZ","canonical_record":{"source":{"id":"2303.06865","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-03-13T05:19:28Z","cross_cats_sorted":["cs.AI","cs.PF"],"title_canon_sha256":"d6cf220d577b076947108f6418a047e01538c181d91d0bba3ae45d3e1c4f5783","abstract_canon_sha256":"01af8b849fbe5aa21c23229c538b9a331283f0caaa4a0fcacce33f83e038dfb9"},"schema_version":"1.0"},"canonical_sha256":"652f5ceb197983ffb36ef388d96cfad61984cbe1046cd938d58c3281faa6db7e","source":{"kind":"arxiv","id":"2303.06865","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2303.06865","created_at":"2026-07-05T06:19:39Z"},{"alias_kind":"arxiv_version","alias_value":"2303.06865v2","created_at":"2026-07-05T06:19:39Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.06865","created_at":"2026-07-05T06:19:39Z"},{"alias_kind":"pith_short_12","alias_value":"MUXVZ2YZPGB7","created_at":"2026-07-05T06:19:39Z"},{"alias_kind":"pith_short_16","alias_value":"MUXVZ2YZPGB77M3O","created_at":"2026-07-05T06:19:39Z"},{"alias_kind":"pith_short_8","alias_value":"MUXVZ2YZ","created_at":"2026-07-05T06:19:39Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:MUXVZ2YZPGB77M3O6OENS3H22Y","target":"record","payload":{"canonical_record":{"source":{"id":"2303.06865","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-03-13T05:19:28Z","cross_cats_sorted":["cs.AI","cs.PF"],"title_canon_sha256":"d6cf220d577b076947108f6418a047e01538c181d91d0bba3ae45d3e1c4f5783","abstract_canon_sha256":"01af8b849fbe5aa21c23229c538b9a331283f0caaa4a0fcacce33f83e038dfb9"},"schema_version":"1.0"},"canonical_sha256":"652f5ceb197983ffb36ef388d96cfad61984cbe1046cd938d58c3281faa6db7e","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:19:39.855648Z","signature_b64":"7KziZ/NU3vgAV5PlLmaBhO7x7g1EX4TtQN+z8OPeUU1uzLxccygoBPI431+rukAgaPFAcg1AfyqDE1qHcK31AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"652f5ceb197983ffb36ef388d96cfad61984cbe1046cd938d58c3281faa6db7e","last_reissued_at":"2026-07-05T06:19:39.855124Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:19:39.855124Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2303.06865","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T06:19:39Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TrkAwC3KulGQrzyaEVIgQH2pBqCI1oFDgITPMR6OioSe8R7hh1rHVNH2zietB826+kImAVHb0+eP4GBAPkzDBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T16:47:22.788902Z"},"content_sha256":"d0e75145a6142f4a4d2e21c8a70ff9e4e5d7942b4838801efd0f5d6c054fabdb","schema_version":"1.0","event_id":"sha256:d0e75145a6142f4a4d2e21c8a70ff9e4e5d7942b4838801efd0f5d6c054fabdb"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:MUXVZ2YZPGB77M3O6OENS3H22Y","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"FlexGen: High-Throughput Generative Inference of Large Language Models with a Single GPU","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.PF"],"primary_cat":"cs.LG","authors_text":"Beidi Chen, Binhang Yuan, Ce Zhang, Christopher R\\'e, Clark Barrett, Daniel Y. Fu, Ion Stoica, Joseph E. Gonzalez, Lianmin Zheng, Max Ryabinin, Percy Liang, Ying Sheng, Zhiqiang Xie, Zhuohan Li","submitted_at":"2023-03-13T05:19:28Z","abstract_excerpt":"The high computational and memory requirements of large language model (LLM) inference make it feasible only with multiple high-end accelerators. Motivated by the emerging demand for latency-insensitive tasks with batched processing, this paper initiates the study of high-throughput LLM inference using limited resources, such as a single commodity GPU. We present FlexGen, a high-throughput generation engine for running LLMs with limited GPU memory. FlexGen can be flexibly configured under various hardware resource constraints by aggregating memory and computation from the GPU, CPU, and disk. B"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.06865","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/2303.06865/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T06:19:39Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"S/QyFPWRZdqsAY1iXLcCluM5x4tacg2oKHTNpqqR5vxpjsrZ8822hXihgDeGq0Mbbe7UoHgRdhm4bsp5kI4gCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T16:47:22.789469Z"},"content_sha256":"ff4cc6bfe498288d358c469adec636060513d78055b85fc50a45081fe4998c7f","schema_version":"1.0","event_id":"sha256:ff4cc6bfe498288d358c469adec636060513d78055b85fc50a45081fe4998c7f"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/MUXVZ2YZPGB77M3O6OENS3H22Y/bundle.json","state_url":"https://pith.science/pith/MUXVZ2YZPGB77M3O6OENS3H22Y/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/MUXVZ2YZPGB77M3O6OENS3H22Y/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-03T16:47:22Z","links":{"resolver":"https://pith.science/pith/MUXVZ2YZPGB77M3O6OENS3H22Y","bundle":"https://pith.science/pith/MUXVZ2YZPGB77M3O6OENS3H22Y/bundle.json","state":"https://pith.science/pith/MUXVZ2YZPGB77M3O6OENS3H22Y/state.json","well_known_bundle":"https://pith.science/.well-known/pith/MUXVZ2YZPGB77M3O6OENS3H22Y/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:MUXVZ2YZPGB77M3O6OENS3H22Y","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"01af8b849fbe5aa21c23229c538b9a331283f0caaa4a0fcacce33f83e038dfb9","cross_cats_sorted":["cs.AI","cs.PF"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-03-13T05:19:28Z","title_canon_sha256":"d6cf220d577b076947108f6418a047e01538c181d91d0bba3ae45d3e1c4f5783"},"schema_version":"1.0","source":{"id":"2303.06865","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2303.06865","created_at":"2026-07-05T06:19:39Z"},{"alias_kind":"arxiv_version","alias_value":"2303.06865v2","created_at":"2026-07-05T06:19:39Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.06865","created_at":"2026-07-05T06:19:39Z"},{"alias_kind":"pith_short_12","alias_value":"MUXVZ2YZPGB7","created_at":"2026-07-05T06:19:39Z"},{"alias_kind":"pith_short_16","alias_value":"MUXVZ2YZPGB77M3O","created_at":"2026-07-05T06:19:39Z"},{"alias_kind":"pith_short_8","alias_value":"MUXVZ2YZ","created_at":"2026-07-05T06:19:39Z"}],"graph_snapshots":[{"event_id":"sha256:ff4cc6bfe498288d358c469adec636060513d78055b85fc50a45081fe4998c7f","target":"graph","created_at":"2026-07-05T06:19:39Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2303.06865/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The high computational and memory requirements of large language model (LLM) inference make it feasible only with multiple high-end accelerators. Motivated by the emerging demand for latency-insensitive tasks with batched processing, this paper initiates the study of high-throughput LLM inference using limited resources, such as a single commodity GPU. We present FlexGen, a high-throughput generation engine for running LLMs with limited GPU memory. FlexGen can be flexibly configured under various hardware resource constraints by aggregating memory and computation from the GPU, CPU, and disk. B","authors_text":"Beidi Chen, Binhang Yuan, Ce Zhang, Christopher R\\'e, Clark Barrett, Daniel Y. Fu, Ion Stoica, Joseph E. Gonzalez, Lianmin Zheng, Max Ryabinin, Percy Liang, Ying Sheng, Zhiqiang Xie, Zhuohan Li","cross_cats":["cs.AI","cs.PF"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-03-13T05:19:28Z","title":"FlexGen: High-Throughput Generative Inference of Large Language Models with a Single GPU"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.06865","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:d0e75145a6142f4a4d2e21c8a70ff9e4e5d7942b4838801efd0f5d6c054fabdb","target":"record","created_at":"2026-07-05T06:19:39Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"01af8b849fbe5aa21c23229c538b9a331283f0caaa4a0fcacce33f83e038dfb9","cross_cats_sorted":["cs.AI","cs.PF"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-03-13T05:19:28Z","title_canon_sha256":"d6cf220d577b076947108f6418a047e01538c181d91d0bba3ae45d3e1c4f5783"},"schema_version":"1.0","source":{"id":"2303.06865","kind":"arxiv","version":2}},"canonical_sha256":"652f5ceb197983ffb36ef388d96cfad61984cbe1046cd938d58c3281faa6db7e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"652f5ceb197983ffb36ef388d96cfad61984cbe1046cd938d58c3281faa6db7e","first_computed_at":"2026-07-05T06:19:39.855124Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:19:39.855124Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"7KziZ/NU3vgAV5PlLmaBhO7x7g1EX4TtQN+z8OPeUU1uzLxccygoBPI431+rukAgaPFAcg1AfyqDE1qHcK31AQ==","signature_status":"signed_v1","signed_at":"2026-07-05T06:19:39.855648Z","signed_message":"canonical_sha256_bytes"},"source_id":"2303.06865","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d0e75145a6142f4a4d2e21c8a70ff9e4e5d7942b4838801efd0f5d6c054fabdb","sha256:ff4cc6bfe498288d358c469adec636060513d78055b85fc50a45081fe4998c7f"],"state_sha256":"26f56a27fc24f4ed0b5205b5d6b4b30fed778d599d11e3ecf5cdb98c5d4a57e2"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"fltqTXW2vG0Ly4NpwAA5IYMmUqDDJU3Jy0CuiP10SBlOhtVvrr/+mh+4zOjqrf/RQRElZ31/Xf17h7EkDg3ZDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-03T16:47:22.795463Z","bundle_sha256":"2c27d196c72c31f7e9c00aa6d5729d37c7fc969c5d5bbc20d2c30a243e1ba7af"}}