{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:HFZNVY5BROR5P5X77SCIA34BBT","short_pith_number":"pith:HFZNVY5B","schema_version":"1.0","canonical_sha256":"3972dae3a18ba3d7f6fffc84806f810cfb775082b8b48efc2236ca15b0788afc","source":{"kind":"arxiv","id":"2605.16259","version":1},"attestation_state":"computed","paper":{"title":"Systematic Optimization of Real-Time Diffusion Model Inference on Apple M3 Ultra","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.DC"],"primary_cat":"cs.LG","authors_text":"Yoichi Ochiai","submitted_at":"2026-02-10T20:40:45Z","abstract_excerpt":"While real-time image generation using diffusion models has advanced rapidly on NVIDIA GPUs, systematic optimization research on non-CUDA platforms such as Apple Silicon remains extremely limited. In this study, we conducted comprehensive optimization experiments across 10 phases targeting the Apple M3 Ultra (60-core GPU, 512 GB unified memory) with the goal of achieving real-time camera img2img transformation. We explored a wide range of techniques including CoreML conversion, quantization, Token Merging, Neural Engine utilization, compact model exploration, frame interpolation, kNN search-ba"},"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":"2605.16259","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2026-02-10T20:40:45Z","cross_cats_sorted":["cs.AI","cs.DC"],"title_canon_sha256":"8fe35820e6fa199e69e2d082f2b52b3ae96fe72266a1ffd781ffb2897c89eeb6","abstract_canon_sha256":"675983778019f6b08510368cf6b64344c052e8c055e96c3f945f00cab6dfd891"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:02:13.877132Z","signature_b64":"LlvyjreQ/GwfdRvtDjuxEARMSsln7l3vefkMeQLiUDhRPfWHB34fBDI+wCgbUlteNpZe9QHeHtm/PDmmbhFVCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3972dae3a18ba3d7f6fffc84806f810cfb775082b8b48efc2236ca15b0788afc","last_reissued_at":"2026-05-20T00:02:13.876420Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:02:13.876420Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Systematic Optimization of Real-Time Diffusion Model Inference on Apple M3 Ultra","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.DC"],"primary_cat":"cs.LG","authors_text":"Yoichi Ochiai","submitted_at":"2026-02-10T20:40:45Z","abstract_excerpt":"While real-time image generation using diffusion models has advanced rapidly on NVIDIA GPUs, systematic optimization research on non-CUDA platforms such as Apple Silicon remains extremely limited. In this study, we conducted comprehensive optimization experiments across 10 phases targeting the Apple M3 Ultra (60-core GPU, 512 GB unified memory) with the goal of achieving real-time camera img2img transformation. We explored a wide range of techniques including CoreML conversion, quantization, Token Merging, Neural Engine utilization, compact model exploration, frame interpolation, kNN search-ba"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.16259","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/2605.16259/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":"2605.16259","created_at":"2026-05-20T00:02:13.876547+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.16259v1","created_at":"2026-05-20T00:02:13.876547+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.16259","created_at":"2026-05-20T00:02:13.876547+00:00"},{"alias_kind":"pith_short_12","alias_value":"HFZNVY5BROR5","created_at":"2026-05-20T00:02:13.876547+00:00"},{"alias_kind":"pith_short_16","alias_value":"HFZNVY5BROR5P5X7","created_at":"2026-05-20T00:02:13.876547+00:00"},{"alias_kind":"pith_short_8","alias_value":"HFZNVY5B","created_at":"2026-05-20T00:02:13.876547+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/HFZNVY5BROR5P5X77SCIA34BBT","json":"https://pith.science/pith/HFZNVY5BROR5P5X77SCIA34BBT.json","graph_json":"https://pith.science/api/pith-number/HFZNVY5BROR5P5X77SCIA34BBT/graph.json","events_json":"https://pith.science/api/pith-number/HFZNVY5BROR5P5X77SCIA34BBT/events.json","paper":"https://pith.science/paper/HFZNVY5B"},"agent_actions":{"view_html":"https://pith.science/pith/HFZNVY5BROR5P5X77SCIA34BBT","download_json":"https://pith.science/pith/HFZNVY5BROR5P5X77SCIA34BBT.json","view_paper":"https://pith.science/paper/HFZNVY5B","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.16259&json=true","fetch_graph":"https://pith.science/api/pith-number/HFZNVY5BROR5P5X77SCIA34BBT/graph.json","fetch_events":"https://pith.science/api/pith-number/HFZNVY5BROR5P5X77SCIA34BBT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HFZNVY5BROR5P5X77SCIA34BBT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HFZNVY5BROR5P5X77SCIA34BBT/action/storage_attestation","attest_author":"https://pith.science/pith/HFZNVY5BROR5P5X77SCIA34BBT/action/author_attestation","sign_citation":"https://pith.science/pith/HFZNVY5BROR5P5X77SCIA34BBT/action/citation_signature","submit_replication":"https://pith.science/pith/HFZNVY5BROR5P5X77SCIA34BBT/action/replication_record"}},"created_at":"2026-05-20T00:02:13.876547+00:00","updated_at":"2026-05-20T00:02:13.876547+00:00"}