{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:DDMY6GIIR7VE4D2EMSPCHIKBNG","short_pith_number":"pith:DDMY6GII","canonical_record":{"source":{"id":"2506.04645","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-05T05:28:09Z","cross_cats_sorted":["cs.DC"],"title_canon_sha256":"343372395adf22dd0a13c819876cea6d6c8b3d5e741a2472d6634103da8a0db1","abstract_canon_sha256":"4d71672d303eba2a680f9191c0a3fc761e1b27b2a95b793a51227909f7f46c4d"},"schema_version":"1.0"},"canonical_sha256":"18d98f19088fea4e0f44649e23a14169a600211a97a486a5071ab4184821424b","source":{"kind":"arxiv","id":"2506.04645","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.04645","created_at":"2026-07-05T11:16:18Z"},{"alias_kind":"arxiv_version","alias_value":"2506.04645v1","created_at":"2026-07-05T11:16:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.04645","created_at":"2026-07-05T11:16:18Z"},{"alias_kind":"pith_short_12","alias_value":"DDMY6GIIR7VE","created_at":"2026-07-05T11:16:18Z"},{"alias_kind":"pith_short_16","alias_value":"DDMY6GIIR7VE4D2E","created_at":"2026-07-05T11:16:18Z"},{"alias_kind":"pith_short_8","alias_value":"DDMY6GII","created_at":"2026-07-05T11:16:18Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:DDMY6GIIR7VE4D2EMSPCHIKBNG","target":"record","payload":{"canonical_record":{"source":{"id":"2506.04645","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-05T05:28:09Z","cross_cats_sorted":["cs.DC"],"title_canon_sha256":"343372395adf22dd0a13c819876cea6d6c8b3d5e741a2472d6634103da8a0db1","abstract_canon_sha256":"4d71672d303eba2a680f9191c0a3fc761e1b27b2a95b793a51227909f7f46c4d"},"schema_version":"1.0"},"canonical_sha256":"18d98f19088fea4e0f44649e23a14169a600211a97a486a5071ab4184821424b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:16:18.434649Z","signature_b64":"eMG0jtdsyxCM1TSmgOMsK0yfDmCLEZ0AHnlaPrI5rdZ9w06ueRPm7Gn7sWYdNefgDeKRWMr6LsT7MkxhF78pBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"18d98f19088fea4e0f44649e23a14169a600211a97a486a5071ab4184821424b","last_reissued_at":"2026-07-05T11:16:18.434209Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:16:18.434209Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.04645","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:16:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nfTkO0HfdZlvSaiDEkJJP/cpUdF8hTKcvgY29lzWA7iznYkWJMoZzBigx2s6aulJzUfGxDkNJYPc0ok1NMLXBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T11:43:34.721019Z"},"content_sha256":"4f9b86d39768eed13b0ef39d1d3995a40f5ed504e5a69252d0e94d55d89cdd42","schema_version":"1.0","event_id":"sha256:4f9b86d39768eed13b0ef39d1d3995a40f5ed504e5a69252d0e94d55d89cdd42"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:DDMY6GIIR7VE4D2EMSPCHIKBNG","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Inference economics of language models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.LG","authors_text":"Ege Erdil","submitted_at":"2025-06-05T05:28:09Z","abstract_excerpt":"We develop a theoretical model that addresses the economic trade-off between cost per token versus serial token generation speed when deploying LLMs for inference at scale. Our model takes into account arithmetic, memory bandwidth, network bandwidth and latency constraints; and optimizes over different parallelism setups and batch sizes to find the ones that optimize serial inference speed at a given cost per token. We use the model to compute Pareto frontiers of serial speed versus cost per token for popular language models."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.04645","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/2506.04645/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":"2026-07-05T11:16:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"izmyhj+TPmYYtloosg78/86V2JDGSIgDLJ/zTtIhq6sKCS/WGvBKYPfJy1f471e+6gFQpq4DW4cK6BGbL7XICA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T11:43:34.721578Z"},"content_sha256":"f7edbe06dcd30f7d83ff634e8bd26bb8965b672ac6820a06a34fcead37502cbb","schema_version":"1.0","event_id":"sha256:f7edbe06dcd30f7d83ff634e8bd26bb8965b672ac6820a06a34fcead37502cbb"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/DDMY6GIIR7VE4D2EMSPCHIKBNG/bundle.json","state_url":"https://pith.science/pith/DDMY6GIIR7VE4D2EMSPCHIKBNG/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/DDMY6GIIR7VE4D2EMSPCHIKBNG/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":"2026-08-05T11:43:34Z","links":{"resolver":"https://pith.science/pith/DDMY6GIIR7VE4D2EMSPCHIKBNG","bundle":"https://pith.science/pith/DDMY6GIIR7VE4D2EMSPCHIKBNG/bundle.json","state":"https://pith.science/pith/DDMY6GIIR7VE4D2EMSPCHIKBNG/state.json","well_known_bundle":"https://pith.science/.well-known/pith/DDMY6GIIR7VE4D2EMSPCHIKBNG/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:DDMY6GIIR7VE4D2EMSPCHIKBNG","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":"4d71672d303eba2a680f9191c0a3fc761e1b27b2a95b793a51227909f7f46c4d","cross_cats_sorted":["cs.DC"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-05T05:28:09Z","title_canon_sha256":"343372395adf22dd0a13c819876cea6d6c8b3d5e741a2472d6634103da8a0db1"},"schema_version":"1.0","source":{"id":"2506.04645","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.04645","created_at":"2026-07-05T11:16:18Z"},{"alias_kind":"arxiv_version","alias_value":"2506.04645v1","created_at":"2026-07-05T11:16:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.04645","created_at":"2026-07-05T11:16:18Z"},{"alias_kind":"pith_short_12","alias_value":"DDMY6GIIR7VE","created_at":"2026-07-05T11:16:18Z"},{"alias_kind":"pith_short_16","alias_value":"DDMY6GIIR7VE4D2E","created_at":"2026-07-05T11:16:18Z"},{"alias_kind":"pith_short_8","alias_value":"DDMY6GII","created_at":"2026-07-05T11:16:18Z"}],"graph_snapshots":[{"event_id":"sha256:f7edbe06dcd30f7d83ff634e8bd26bb8965b672ac6820a06a34fcead37502cbb","target":"graph","created_at":"2026-07-05T11:16:18Z","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/2506.04645/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We develop a theoretical model that addresses the economic trade-off between cost per token versus serial token generation speed when deploying LLMs for inference at scale. Our model takes into account arithmetic, memory bandwidth, network bandwidth and latency constraints; and optimizes over different parallelism setups and batch sizes to find the ones that optimize serial inference speed at a given cost per token. We use the model to compute Pareto frontiers of serial speed versus cost per token for popular language models.","authors_text":"Ege Erdil","cross_cats":["cs.DC"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-05T05:28:09Z","title":"Inference economics of language models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.04645","kind":"arxiv","version":1},"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":"sha256:4f9b86d39768eed13b0ef39d1d3995a40f5ed504e5a69252d0e94d55d89cdd42","target":"record","created_at":"2026-07-05T11:16:18Z","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":"4d71672d303eba2a680f9191c0a3fc761e1b27b2a95b793a51227909f7f46c4d","cross_cats_sorted":["cs.DC"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-05T05:28:09Z","title_canon_sha256":"343372395adf22dd0a13c819876cea6d6c8b3d5e741a2472d6634103da8a0db1"},"schema_version":"1.0","source":{"id":"2506.04645","kind":"arxiv","version":1}},"canonical_sha256":"18d98f19088fea4e0f44649e23a14169a600211a97a486a5071ab4184821424b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"18d98f19088fea4e0f44649e23a14169a600211a97a486a5071ab4184821424b","first_computed_at":"2026-07-05T11:16:18.434209Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:16:18.434209Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"eMG0jtdsyxCM1TSmgOMsK0yfDmCLEZ0AHnlaPrI5rdZ9w06ueRPm7Gn7sWYdNefgDeKRWMr6LsT7MkxhF78pBw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:16:18.434649Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.04645","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4f9b86d39768eed13b0ef39d1d3995a40f5ed504e5a69252d0e94d55d89cdd42","sha256:f7edbe06dcd30f7d83ff634e8bd26bb8965b672ac6820a06a34fcead37502cbb"],"state_sha256":"d55f2a7379e600f3708f6390382de1f1f6f08c30f8ae1daa15bcc3e31ebedaad"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"u3fNZkMvj+SScdSDzGyXpYSC5QsNlUhv3iJ+CCTX10lHU0h7YJaYJMHjalqSq0K01m9FiHzj/c1J0pyBz2QvCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T11:43:34.725202Z","bundle_sha256":"2534fbc027da8dbab770be8989dc64a08ed9cc70d88070b7b3a5a86a3c7272e8"}}