{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:WQVS3NZVBCGQL23OXG3YERNIEA","short_pith_number":"pith:WQVS3NZV","schema_version":"1.0","canonical_sha256":"b42b2db735088d05eb6eb9b78245a82038e2d80c6504800d7168906561968bfc","source":{"kind":"arxiv","id":"2310.03003","version":1},"attestation_state":"computed","paper":{"title":"From Words to Watts: Benchmarking the Energy Costs of Large Language Model Inference","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.CL","authors_text":"Adam Michaleas, Baolin Li, Dan Zhao, Devesh Tiwari, Jeremy Kepner, Joseph McDonald, Michael Jones, Siddharth Samsi, Vijay Gadepally, William Bergeron","submitted_at":"2023-10-04T17:41:59Z","abstract_excerpt":"Large language models (LLMs) have exploded in popularity due to their new generative capabilities that go far beyond prior state-of-the-art. These technologies are increasingly being leveraged in various domains such as law, finance, and medicine. However, these models carry significant computational challenges, especially the compute and energy costs required for inference. Inference energy costs already receive less attention than the energy costs of training LLMs -- despite how often these large models are called on to conduct inference in reality (e.g., ChatGPT). As these state-of-the-art "},"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":"2310.03003","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-10-04T17:41:59Z","cross_cats_sorted":["cs.DC"],"title_canon_sha256":"e4b0203469205d893a313bbab16690d72964259d83245351e208b0951d38a7e6","abstract_canon_sha256":"40e94cfcbd5eae3d078586a2858c437fea05f9b8bcb8b27c8a946610f4ea403a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:57:16.413296Z","signature_b64":"s86r5ACCtPyzyn/IdqYu1bYQ8rop1d9ZAXawpiXP0+t/SRZUdqYzp77N3IuZk3C3ch054TyK500CHLtX03WQBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b42b2db735088d05eb6eb9b78245a82038e2d80c6504800d7168906561968bfc","last_reissued_at":"2026-07-05T06:57:16.412796Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:57:16.412796Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"From Words to Watts: Benchmarking the Energy Costs of Large Language Model Inference","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.CL","authors_text":"Adam Michaleas, Baolin Li, Dan Zhao, Devesh Tiwari, Jeremy Kepner, Joseph McDonald, Michael Jones, Siddharth Samsi, Vijay Gadepally, William Bergeron","submitted_at":"2023-10-04T17:41:59Z","abstract_excerpt":"Large language models (LLMs) have exploded in popularity due to their new generative capabilities that go far beyond prior state-of-the-art. These technologies are increasingly being leveraged in various domains such as law, finance, and medicine. However, these models carry significant computational challenges, especially the compute and energy costs required for inference. Inference energy costs already receive less attention than the energy costs of training LLMs -- despite how often these large models are called on to conduct inference in reality (e.g., ChatGPT). As these state-of-the-art "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.03003","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/2310.03003/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":"2310.03003","created_at":"2026-07-05T06:57:16.412854+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.03003v1","created_at":"2026-07-05T06:57:16.412854+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.03003","created_at":"2026-07-05T06:57:16.412854+00:00"},{"alias_kind":"pith_short_12","alias_value":"WQVS3NZVBCGQ","created_at":"2026-07-05T06:57:16.412854+00:00"},{"alias_kind":"pith_short_16","alias_value":"WQVS3NZVBCGQL23O","created_at":"2026-07-05T06:57:16.412854+00:00"},{"alias_kind":"pith_short_8","alias_value":"WQVS3NZV","created_at":"2026-07-05T06:57:16.412854+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":9,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.26749","citing_title":"Structure Before Collapse: Transient semantic geometry in next-token prediction","ref_index":98,"is_internal_anchor":false},{"citing_arxiv_id":"2504.13217","citing_title":"Sustainability via LLM Right-sizing","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2511.04776","citing_title":"Quantifying the Climate Risk of Generative AI: Region-Aware Carbon Accounting with G-TRACE and the AI Sustainability Pyramid","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17159","citing_title":"MADP: A Multi-Agent Pipeline for Sustainable Document Processing with Human-in-the-Loop","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2511.04776","citing_title":"Quantifying the Climate Risk of Generative AI: Region-Aware Carbon Accounting with G-TRACE and the AI Sustainability Pyramid","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2602.20669","citing_title":"Integrating Domain-Specialized Language Models with AI Measurement Tools for Deterministic Atomic-Resolution Experimentation","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2603.00989","citing_title":"Sustainable Code Generation Using Large Language Models: A Systematic Literature Review","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11733","citing_title":"Position: LLM Inference Should Be Evaluated as Energy-to-Token Production","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2604.04745","citing_title":"The Energy Cost of Execution-Idle in GPU Clusters","ref_index":48,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WQVS3NZVBCGQL23OXG3YERNIEA","json":"https://pith.science/pith/WQVS3NZVBCGQL23OXG3YERNIEA.json","graph_json":"https://pith.science/api/pith-number/WQVS3NZVBCGQL23OXG3YERNIEA/graph.json","events_json":"https://pith.science/api/pith-number/WQVS3NZVBCGQL23OXG3YERNIEA/events.json","paper":"https://pith.science/paper/WQVS3NZV"},"agent_actions":{"view_html":"https://pith.science/pith/WQVS3NZVBCGQL23OXG3YERNIEA","download_json":"https://pith.science/pith/WQVS3NZVBCGQL23OXG3YERNIEA.json","view_paper":"https://pith.science/paper/WQVS3NZV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.03003&json=true","fetch_graph":"https://pith.science/api/pith-number/WQVS3NZVBCGQL23OXG3YERNIEA/graph.json","fetch_events":"https://pith.science/api/pith-number/WQVS3NZVBCGQL23OXG3YERNIEA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WQVS3NZVBCGQL23OXG3YERNIEA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WQVS3NZVBCGQL23OXG3YERNIEA/action/storage_attestation","attest_author":"https://pith.science/pith/WQVS3NZVBCGQL23OXG3YERNIEA/action/author_attestation","sign_citation":"https://pith.science/pith/WQVS3NZVBCGQL23OXG3YERNIEA/action/citation_signature","submit_replication":"https://pith.science/pith/WQVS3NZVBCGQL23OXG3YERNIEA/action/replication_record"}},"created_at":"2026-07-05T06:57:16.412854+00:00","updated_at":"2026-07-05T06:57:16.412854+00:00"}