{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:DZZSXEEW7HPDEZDAB754S4UEUI","short_pith_number":"pith:DZZSXEEW","schema_version":"1.0","canonical_sha256":"1e732b9096f9de3264600ffbc97284a224dffdd907bed410480a175da93e17f1","source":{"kind":"arxiv","id":"2605.19275","version":1},"attestation_state":"computed","paper":{"title":"Open-Weight LLMs Are Often Competitive with Commercial APIs for Political Science Text Classification","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"stat.AP","authors_text":"Hanno Hilbig","submitted_at":"2026-05-19T02:46:35Z","abstract_excerpt":"Can researchers use local open-weight models instead of commercial APIs for LLM text classification? Local models avoid marginal API charges, keep data on the researcher's machine, and make exact model versions easier to preserve. I benchmark five local models against four commercial API models on 34 political science classification tasks. Local models are often competitive, especially on simpler tasks. In a task-specific oracle comparison, local models match or exceed API performance on 9 tasks; on average, the best API model exceeds the best local model by 0.015 F1. The four strongest observ"},"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":"2605.19275","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.AP","submitted_at":"2026-05-19T02:46:35Z","cross_cats_sorted":[],"title_canon_sha256":"e796cccb97d8160d2b325b450a5b87d0d78687765cbf85cec3ce8c8817b26ade","abstract_canon_sha256":"f30cd3d47841d67750c89c353328358327b4713d310a615315512908ac66b206"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T01:05:37.078666Z","signature_b64":"nLic/vvZFBBeeic8C9dHGtPdp+YNAyCzARhzKPkFkY8PHnR+X6aPueJR5PVaTpuEyYO2I1Xa8Xraf+C+RcoVDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1e732b9096f9de3264600ffbc97284a224dffdd907bed410480a175da93e17f1","last_reissued_at":"2026-05-20T01:05:37.078128Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T01:05:37.078128Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Open-Weight LLMs Are Often Competitive with Commercial APIs for Political Science Text Classification","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"stat.AP","authors_text":"Hanno Hilbig","submitted_at":"2026-05-19T02:46:35Z","abstract_excerpt":"Can researchers use local open-weight models instead of commercial APIs for LLM text classification? Local models avoid marginal API charges, keep data on the researcher's machine, and make exact model versions easier to preserve. I benchmark five local models against four commercial API models on 34 political science classification tasks. Local models are often competitive, especially on simpler tasks. In a task-specific oracle comparison, local models match or exceed API performance on 9 tasks; on average, the best API model exceeds the best local model by 0.015 F1. The four strongest observ"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.19275","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/2605.19275/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":"2605.19275","created_at":"2026-05-20T01:05:37.078225+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.19275v1","created_at":"2026-05-20T01:05:37.078225+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.19275","created_at":"2026-05-20T01:05:37.078225+00:00"},{"alias_kind":"pith_short_12","alias_value":"DZZSXEEW7HPD","created_at":"2026-05-20T01:05:37.078225+00:00"},{"alias_kind":"pith_short_16","alias_value":"DZZSXEEW7HPDEZDA","created_at":"2026-05-20T01:05:37.078225+00:00"},{"alias_kind":"pith_short_8","alias_value":"DZZSXEEW","created_at":"2026-05-20T01:05:37.078225+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DZZSXEEW7HPDEZDAB754S4UEUI","json":"https://pith.science/pith/DZZSXEEW7HPDEZDAB754S4UEUI.json","graph_json":"https://pith.science/api/pith-number/DZZSXEEW7HPDEZDAB754S4UEUI/graph.json","events_json":"https://pith.science/api/pith-number/DZZSXEEW7HPDEZDAB754S4UEUI/events.json","paper":"https://pith.science/paper/DZZSXEEW"},"agent_actions":{"view_html":"https://pith.science/pith/DZZSXEEW7HPDEZDAB754S4UEUI","download_json":"https://pith.science/pith/DZZSXEEW7HPDEZDAB754S4UEUI.json","view_paper":"https://pith.science/paper/DZZSXEEW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.19275&json=true","fetch_graph":"https://pith.science/api/pith-number/DZZSXEEW7HPDEZDAB754S4UEUI/graph.json","fetch_events":"https://pith.science/api/pith-number/DZZSXEEW7HPDEZDAB754S4UEUI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DZZSXEEW7HPDEZDAB754S4UEUI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DZZSXEEW7HPDEZDAB754S4UEUI/action/storage_attestation","attest_author":"https://pith.science/pith/DZZSXEEW7HPDEZDAB754S4UEUI/action/author_attestation","sign_citation":"https://pith.science/pith/DZZSXEEW7HPDEZDAB754S4UEUI/action/citation_signature","submit_replication":"https://pith.science/pith/DZZSXEEW7HPDEZDAB754S4UEUI/action/replication_record"}},"created_at":"2026-05-20T01:05:37.078225+00:00","updated_at":"2026-05-20T01:05:37.078225+00:00"}