{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:LGDTXFMXU7D4J4HUIDUY6RNOFQ","short_pith_number":"pith:LGDTXFMX","schema_version":"1.0","canonical_sha256":"59873b9597a7c7c4f0f440e98f45ae2c184cb124bcd8541b3b9108744f5f1399","source":{"kind":"arxiv","id":"2606.24031","version":1},"attestation_state":"computed","paper":{"title":"A 35B Hybrid-Attention Mixture-of-Experts Model on a 6GB 2011 GPU: Hand-Written 4-bit CUDA Inference for Fermi","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cond-mat.other","authors_text":"A. C. Opus, J. Q. Lu","submitted_at":"2026-06-23T00:13:16Z","abstract_excerpt":"We report end-to-end inference of \\textbf{Qwen3.6-35B-A3B} -- a 35-billion-parameter, $\\sim$3B-active Mixture-of-Experts (MoE) model with a hybrid gated-delta-net / full-attention backbone -- on a \\textbf{2011 NVIDIA Tesla C2075} (Fermi, compute capability \\smtwenty, 6\\,GB), a GPU that predates tensor cores, native FP16 arithmetic, the \\texttt{DP4A} integer dot-product instruction, and support in every modern CUDA toolchain. Because the 4-bit model ($\\approx$10.5\\,GB) is roughly twice the device memory, we adopt a \\emph{hybrid} execution strategy: the GPU performs batched prompt \\emph{prefill}"},"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":"2606.24031","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cond-mat.other","submitted_at":"2026-06-23T00:13:16Z","cross_cats_sorted":[],"title_canon_sha256":"4918290269f4c96e0f7c73b05df3d2f1393298a94937089920679859e250fc2c","abstract_canon_sha256":"f74ea153b4b6975f53291be34650a4d96d82cff8877a663b07cf7cb64f1d81d6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-24T01:14:38.583833Z","signature_b64":"SzWnUt+U49FHup9jGuZ8Df8G/6rmsLsdDsaxuHjQ4YiFAgmaBqHHkknJL5CKpGcwkPkGR+iQ3r3vS6WQpiuBAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"59873b9597a7c7c4f0f440e98f45ae2c184cb124bcd8541b3b9108744f5f1399","last_reissued_at":"2026-06-24T01:14:38.583293Z","signature_status":"signed_v1","first_computed_at":"2026-06-24T01:14:38.583293Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A 35B Hybrid-Attention Mixture-of-Experts Model on a 6GB 2011 GPU: Hand-Written 4-bit CUDA Inference for Fermi","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cond-mat.other","authors_text":"A. C. Opus, J. Q. Lu","submitted_at":"2026-06-23T00:13:16Z","abstract_excerpt":"We report end-to-end inference of \\textbf{Qwen3.6-35B-A3B} -- a 35-billion-parameter, $\\sim$3B-active Mixture-of-Experts (MoE) model with a hybrid gated-delta-net / full-attention backbone -- on a \\textbf{2011 NVIDIA Tesla C2075} (Fermi, compute capability \\smtwenty, 6\\,GB), a GPU that predates tensor cores, native FP16 arithmetic, the \\texttt{DP4A} integer dot-product instruction, and support in every modern CUDA toolchain. Because the 4-bit model ($\\approx$10.5\\,GB) is roughly twice the device memory, we adopt a \\emph{hybrid} execution strategy: the GPU performs batched prompt \\emph{prefill}"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.24031","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/2606.24031/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":"2606.24031","created_at":"2026-06-24T01:14:38.583359+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.24031v1","created_at":"2026-06-24T01:14:38.583359+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.24031","created_at":"2026-06-24T01:14:38.583359+00:00"},{"alias_kind":"pith_short_12","alias_value":"LGDTXFMXU7D4","created_at":"2026-06-24T01:14:38.583359+00:00"},{"alias_kind":"pith_short_16","alias_value":"LGDTXFMXU7D4J4HU","created_at":"2026-06-24T01:14:38.583359+00:00"},{"alias_kind":"pith_short_8","alias_value":"LGDTXFMX","created_at":"2026-06-24T01:14:38.583359+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/LGDTXFMXU7D4J4HUIDUY6RNOFQ","json":"https://pith.science/pith/LGDTXFMXU7D4J4HUIDUY6RNOFQ.json","graph_json":"https://pith.science/api/pith-number/LGDTXFMXU7D4J4HUIDUY6RNOFQ/graph.json","events_json":"https://pith.science/api/pith-number/LGDTXFMXU7D4J4HUIDUY6RNOFQ/events.json","paper":"https://pith.science/paper/LGDTXFMX"},"agent_actions":{"view_html":"https://pith.science/pith/LGDTXFMXU7D4J4HUIDUY6RNOFQ","download_json":"https://pith.science/pith/LGDTXFMXU7D4J4HUIDUY6RNOFQ.json","view_paper":"https://pith.science/paper/LGDTXFMX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.24031&json=true","fetch_graph":"https://pith.science/api/pith-number/LGDTXFMXU7D4J4HUIDUY6RNOFQ/graph.json","fetch_events":"https://pith.science/api/pith-number/LGDTXFMXU7D4J4HUIDUY6RNOFQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LGDTXFMXU7D4J4HUIDUY6RNOFQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LGDTXFMXU7D4J4HUIDUY6RNOFQ/action/storage_attestation","attest_author":"https://pith.science/pith/LGDTXFMXU7D4J4HUIDUY6RNOFQ/action/author_attestation","sign_citation":"https://pith.science/pith/LGDTXFMXU7D4J4HUIDUY6RNOFQ/action/citation_signature","submit_replication":"https://pith.science/pith/LGDTXFMXU7D4J4HUIDUY6RNOFQ/action/replication_record"}},"created_at":"2026-06-24T01:14:38.583359+00:00","updated_at":"2026-06-24T01:14:38.583359+00:00"}