{ "slug": "alphafold3", "name": "AlphaFold 3", "kind": "co-folding", "publisher": "Google DeepMind", "tier": "verified", "license_code": { "status": "permissive", "spdx": "Apache-2.0", "custom_license": false, "commercial_use": "yes", "derivatives": "yes", "redistribution": "yes", "license_conditions": [ "notice-required" ], "requires_application": false, "evidence": [ { "location": "repo-license", "excerpt": "Apache License", "url": "https://raw.githubusercontent.com/google-deepmind/alphafold3/main/LICENSE", "snapshot_id": "s03", "char_start": 34, "verified_on": "2026-11-13", "observed": "page", "captured_by": "pipeline" } ] }, "license_weights": { "status": "restricted", "spdx": "LicenseRef-Google-DeepMind-terms", "custom_license": true, "commercial_use": "no", "derivatives": "unstated", "redistribution": "no", "license_conditions": [ "registration-required", "territory-dependent", "notice-required", "attribution" ], "requires_application": false, "evidence": [ { "location": "terms-page", "excerpt": "The AlphaFold 3 model parameters and output are **only** available for", "url": "https://raw.githubusercontent.com/google-deepmind/alphafold3/main/WEIGHTS_TERMS_OF_USE.md", "snapshot_id": "s12", "char_start": 821, "verified_on": "2026-11-13", "observed": "page", "captured_by": "pipeline" }, { "location": "terms-page", "excerpt": "You ***must not* publish or share AlphaFold 3 model parameters**", "url": "https://raw.githubusercontent.com/google-deepmind/alphafold3/main/WEIGHTS_TERMS_OF_USE.md", "snapshot_id": "s12", "char_start": 1752, "verified_on": "2026-11-13", "observed": "page", "captured_by": "pipeline" }, { "location": "terms-page", "excerpt": "The AlphaFold 3 model parameters and output are **only** available for non-commercial use by, or on behalf of, non-commercial organizations (*i.e.*, universities, non-profit organizations and research institutes, educational, journalism and government bodies).", "url": "https://raw.githubusercontent.com/google-deepmind/alphafold3/main/WEIGHTS_TERMS_OF_USE.md", "snapshot_id": "s08", "char_start": null, "verified_on": "2026-11-13", "observed": "page", "captured_by": "pipeline" }, { "location": "model-card-page", "excerpt": "not (i) resident of a embargoed country, (ii) ordinarily resident in a US", "url": "https://raw.githubusercontent.com/google-deepmind/alphafold3/main/WEIGHTS_TERMS_OF_USE.md", "snapshot_id": "s08", "char_start": null, "verified_on": "2026-11-13", "observed": "page", "captured_by": "pipeline" }, { "location": "model-card-page", "excerpt": "Google may at any time suspend or terminate your right to use and, as applicable", "url": "https://raw.githubusercontent.com/google-deepmind/alphafold3/main/WEIGHTS_TERMS_OF_USE.md", "snapshot_id": "s08", "char_start": null, "verified_on": "2026-11-13", "observed": "page", "captured_by": "pipeline" } ] }, "last_verified": "2026-11-14", "schema_version": "2.0", "record_type": "model", "official_urls": { "repository_or_release": "https://github.com/google-deepmind/alphafold3", "weights_page": "https://raw.githubusercontent.com/google-deepmind/alphafold3/main/WEIGHTS_TERMS_OF_USE.md", "paper": "https://arxiv.org/abs/2510.24670", "license_file": "https://raw.githubusercontent.com/google-deepmind/alphafold3/main/LICENSE" }, "version_history": [ { "version": "unspecified", "released_on": "2026-06-09", "code_license": null, "weights_license": null, "changed": "other", "note": null, "evidence": [ { "location": "official-change-notice", "excerpt": "Only the code is re-licensed, the weights license and terms of use stay the same.", "url": "https://github.com/google-deepmind/alphafold3/commit/7b197fe859", "snapshot_id": "s13", "char_start": 2617, "verified_on": "2026-11-13", "observed": "page", "source_work": { "title": "Change the code license from CC BY-NC-SA 4.0 to Apache 2.0 (commit 7b197fe)", "authors_or_org": "Google DeepMind (committed by Augustin-Zidek)", "date": "2026-06-09", "date_as_written": "2026-06-09", "source_type": "official-change-notice", "open_access": true, "doi_or_url": "https://github.com/google-deepmind/alphafold3/commit/7b197fe859" } } ] }, { "version": "AlphaFold 3", "released_on": "2026-07-23", "code_license": null, "weights_license": null, "changed": "other", "note": null, "evidence": [ { "location": "official-change-notice", "excerpt": "request access to the AlphaFold 3 model parameters, please complete", "url": "https://github.com/google-deepmind/alphafold3/commit/dd1a7badb6", "snapshot_id": "s14", "char_start": 3607, "verified_on": "2026-11-13", "observed": "page", "source_work": { "title": "Update the instructions on how to obtain AlphaFold 3 weights (commit dd1a7ba, removed lines)", "authors_or_org": "Google DeepMind (committed by Augustin-Zidek)", "date": "2026-07-23", "date_as_written": "2026-07-23", "source_type": "official-change-notice", "open_access": true, "doi_or_url": "https://github.com/google-deepmind/alphafold3/commit/dd1a7badb6" } } ] } ], "output_restrictions": [ { "kind": "no-training-other-models-on-output", "applies_to": "weights", "statement": { "text": { "en": "Output may not be used to train or create machine learning models or related technology for biomolecular structure prediction similar to AlphaFold 3.", "zh": "不得使用输出结果训练或构建与 AlphaFold 3 类似的生物分子结构预测机器学习模型或相关技术。" }, "evidence": [ { "location": "terms-page", "excerpt": "Use Output to train or create machine learning models or related technology for biomolecular structure prediction similar to AlphaFold 3", "url": "https://raw.githubusercontent.com/google-deepmind/alphafold3/main/WEIGHTS_PROHIBITED_USE_POLICY.md", "snapshot_id": "s05", "char_start": null, "verified_on": "2026-11-13", "observed": "page", "captured_by": "pipeline" } ] } }, { "kind": "non-commercial-output", "applies_to": "weights", "statement": { "text": { "en": "The model parameters and output are available only for non-commercial use by or on behalf of non-commercial organizations such as universities, non-profits, research institutes, educational, journalism and government bodies.", "zh": "模型参数及其输出仅限大学、非营利组织、研究机构、教育机构、新闻机构和政府机构等非商业组织自身或代表其进行非商业使用。" }, "evidence": [ { "location": "terms-page", "excerpt": "The AlphaFold 3 model parameters and output are **only** available for non-commercial use by, or on behalf of, non-commercial organizations (*i.e.*, universities, non-profit organizations and research institutes, educational, journalism and government bodies).", "url": "https://raw.githubusercontent.com/google-deepmind/alphafold3/main/WEIGHTS_TERMS_OF_USE.md", "snapshot_id": "s08", "char_start": null, "verified_on": "2026-11-13", "observed": "page", "captured_by": "pipeline" } ] } }, { "kind": "no-clinical-use", "applies_to": "all", "statement": { "text": { "en": "AlphaFold 3 and its output are for theoretical modeling only and are not intended, validated, or approved for clinical use.", "zh": "AlphaFold 3 及其输出仅用于理论建模，并非为临床用途设计、验证或批准。" }, "evidence": [ { "location": "readme", "excerpt": "AlphaFold 3 and its output are for theoretical modeling only. They are not intended, validated, or approved for clinical use.", "url": "https://raw.githubusercontent.com/google-deepmind/alphafold3/HEAD/README.md", "snapshot_id": "s02", "char_start": null, "verified_on": "2026-11-13", "observed": "page", "captured_by": "pipeline" } ] } }, { "kind": "non-commercial-output", "applies_to": "all", "statement": { "text": { "en": "Output Terms of Use (last modified 2024-11-09) forbid using AF3 output on behalf of commercial organisations or in commercial activities, including research for them.", "zh": "《输出使用条款》（最后修改 2024-11-09）禁止代表商业机构或在商业活动中使用 AF3 输出，包括为其开展的研究。" }, "evidence": [ { "location": "dataset-page", "excerpt": "Output are made available free of charge, for non-commercial use", "url": "https://github.com/google-deepmind/alphafold3/blob/main/OUTPUT_TERMS_OF_USE.md", "snapshot_id": "s15", "char_start": 3076, "verified_on": "2026-11-13", "observed": "page", "source_work": { "title": "AlphaFold 3 Output Terms of Use", "authors_or_org": "Google DeepMind", "date": "2024-11-09", "date_as_written": "2024-11-09", "source_type": "official-page", "open_access": true, "doi_or_url": "https://github.com/google-deepmind/alphafold3/blob/main/OUTPUT_TERMS_OF_USE.md" } } ] } }, { "kind": "no-training-other-models-on-output", "applies_to": "all", "statement": { "text": { "en": "Output must not be used to train or create biomolecular structure prediction models similar to AF3, including via distillation.", "zh": "不得用 AF3 输出训练或构建与 AF3 类似的生物分子结构预测模型，包括通过蒸馏。" }, "evidence": [ { "location": "dataset-page", "excerpt": "biomolecular structure prediction similar to AlphaFold 3 as made available", "url": "https://github.com/google-deepmind/alphafold3/blob/main/OUTPUT_TERMS_OF_USE.md", "snapshot_id": "s15", "char_start": 5970, "verified_on": "2026-11-13", "observed": "page", "source_work": { "title": "AlphaFold 3 Output Terms of Use", "authors_or_org": "Google DeepMind", "date": "2024-11-09", "date_as_written": "2024-11-09", "source_type": "official-page", "open_access": true, "doi_or_url": "https://github.com/google-deepmind/alphafold3/blob/main/OUTPUT_TERMS_OF_USE.md" } } ] } }, { "kind": "non-commercial-output", "applies_to": "weights", "statement": { "text": { "en": "Source sentence: You must not share Output with any commercial organization.", "zh": "来源原句：You must not share Output with any commercial organization." }, "evidence": [ { "location": "model-card-page", "excerpt": "You must not share Output with any commercial organization.", "url": "https://raw.githubusercontent.com/google-deepmind/alphafold3/main/OUTPUT_TERMS_OF_USE.md", "snapshot_id": "s15", "char_start": null, "verified_on": "2026-11-13", "observed": "page", "captured_by": "pipeline" } ] } }, { "kind": "other", "applies_to": "weights", "statement": { "text": { "en": "Source sentence: Distribute Output, or disclose findings arising from using AlphaFold 3 without citing our paper:", "zh": "来源原句：Distribute Output, or disclose findings arising from using AlphaFold 3 without citing our paper:" }, "evidence": [ { "location": "model-card-page", "excerpt": "Distribute Output, or disclose findings arising from using AlphaFold 3 without citing our paper:", "url": "https://raw.githubusercontent.com/google-deepmind/alphafold3/main/WEIGHTS_PROHIBITED_USE_POLICY.md", "snapshot_id": "s05", "char_start": null, "verified_on": "2026-11-13", "observed": "page", "captured_by": "pipeline" } ] } }, { "kind": "other", "applies_to": "weights", "statement": { "text": { "en": "Source sentence: making decisions in domains that affect material or individual rights or", "zh": "来源原句：making decisions in domains that affect material or individual rights or" }, "evidence": [ { "location": "model-card-page", "excerpt": "making decisions in domains that affect material or individual rights or", "url": "https://raw.githubusercontent.com/google-deepmind/alphafold3/main/OUTPUT_TERMS_OF_USE.md", "snapshot_id": "s15", "char_start": null, "verified_on": "2026-11-13", "observed": "page", "captured_by": "pipeline" } ] } } ], "training_data": [ { "resource": "training data", "slug": null, "role": "other", "cutoff_date": null, "cutoff_stated": false, "statement": { "text": { "en": "No structural training data released after 30 September 2021 were used.", "zh": "训练所用结构数据均不晚于 2021-09-30 发布。" }, "evidence": [ { "location": "peer-reviewed-paper", "excerpt": "No structural data used during training were released after 30 September 2021", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC11168924/", "snapshot_id": "s11", "char_start": 53237, "verified_on": "2026-11-13", "observed": "page", "source_work": { "title": "Accurate structure prediction of biomolecular interactions with AlphaFold 3", "authors_or_org": "Abramson J, Adler J, Dunger J, et al. (Nature)", "date": "2024-05-08", "date_as_written": "2024-05-08", "source_type": "peer-reviewed-paper", "open_access": true, "doi_or_url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC11168924/" } } ] } }, { "resource": "training data", "slug": null, "role": "other", "cutoff_date": null, "cutoff_stated": false, "statement": { "text": { "en": "Training data were enriched with AlphaFold-Multimer v2.3 predicted structures (cross-distillation).", "zh": "训练数据中加入了 AlphaFold-Multimer v2.3 的预测结构（交叉蒸馏）。" }, "evidence": [ { "location": "peer-reviewed-paper", "excerpt": "we use a cross-distillation method in which we enrich the training data with structures predicted by AlphaFold-Multimer (v.2.3)", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC11168924/", "snapshot_id": "s11", "char_start": 22274, "verified_on": "2026-11-13", "observed": "page", "source_work": { "title": "Accurate structure prediction of biomolecular interactions with AlphaFold 3", "authors_or_org": "Abramson J, Adler J, Dunger J, et al. (Nature)", "date": "2024-05-08", "date_as_written": "2024-05-08", "source_type": "peer-reviewed-paper", "open_access": true, "doi_or_url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC11168924/" } } ] } } ], "benchmark_overlap": [ { "benchmark": "PoseBusters", "benchmark_slug": null, "finding": "cutoff-after-benchmark", "citation": { "reference": "https://pmc.ncbi.nlm.nih.gov/articles/PMC11168924/", "statement": { "text": { "en": "Because the standard cutoff is in 2021 and PoseBusters contains structures released in 2021 or later, AF3 authors trained a separate earlier-cutoff model for PoseBusters evaluation.", "zh": "由于标准截止日期在 2021 年，而 PoseBusters 含 2021 年及以后发布的结构，AF3 作者为 PoseBusters 评测另训了一个截止更早的模型。" }, "evidence": [ { "location": "peer-reviewed-paper", "excerpt": "As our standard training cut-off date is in 2021, we trained a separate AF3 model with an earlier training-set cutoff ( Methods ).", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC11168924/", "snapshot_id": "s11", "char_start": 28438, "verified_on": "2026-11-13", "observed": "page", "source_work": { "title": "Accurate structure prediction of biomolecular interactions with AlphaFold 3", "authors_or_org": "Abramson J, Adler J, Dunger J, et al. (Nature)", "date": "2024-05-08", "date_as_written": "2024-05-08", "source_type": "peer-reviewed-paper", "open_access": true, "doi_or_url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC11168924/" } } ] } } }, { "benchmark": "benchmark", "benchmark_slug": null, "finding": "overlap-reported", "citation": { "reference": "https://pmc.ncbi.nlm.nih.gov/articles/PMC12501370/", "statement": { "text": { "en": "Masters, Mahmoud and Lill tested AF3, RoseTTAFold All-Atom, Chai-1 and Boltz-1 with adversarial binding-site edits: with all binding-site residues mutated to glycine, all four still predicted the same ATP–CDK2 binding mode, which the authors read as possible overfitting to training-data features.", "zh": "Masters、Mahmoud、Lill 用对抗性结合位点改造测试 AF3、RoseTTAFold All-Atom、Chai-1、Boltz-1：把结合位点残基全部突变为甘氨酸后，四个模型仍预测相同的 ATP–CDK2 结合模式，作者认为可能是对训练数据特征的过拟合。" }, "evidence": [ { "location": "peer-reviewed-paper", "excerpt": "All four co-folding models continue to predict the ATP-CDK2 complex with the same binding mode, despite the loss of all major side-chain interactions.", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC12501370/", "snapshot_id": "s18", "char_start": 15315, "verified_on": "2026-11-13", "observed": "page", "source_work": { "title": "Investigating whether deep learning models for co-folding learn the physics of protein-ligand interactions", "authors_or_org": "Masters MR, Mahmoud AH, Lill MA (Nature Communications)", "date": "2025-10-06", "date_as_written": "2025-10-06", "source_type": "peer-reviewed-paper", "open_access": true, "doi_or_url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC12501370/" } } ] } } }, { "benchmark": "Runs N' Poses", "benchmark_slug": null, "finding": "memorisation-reported", "citation": { "reference": "https://github.com/plinder-org/runs-n-poses", "statement": { "text": { "en": "Runs N' Poses is designed as a zero-shot benchmark for methods with a 30 September 2021 structural cutoff (AF3's cutoff); its authors conclude current co-folding methods largely memorise training ligand poses (the abstract does not name the four methods; the names were not verified).", "zh": "Runs N' Poses 是为结构截止日期为 2021-09-30（即 AF3 的截止日期）的方法设计的零样本基准；作者结论是现有共折叠方法很大程度上记忆训练集配体构象（摘要未列出所评四个方法的名称，未核实）。" }, "evidence": [ { "location": "dataset-page", "excerpt": "This is a zero-shot benchmark, provided that your method uses a structural training cutoff of 30 September 2021.", "url": "https://github.com/plinder-org/runs-n-poses", "snapshot_id": "s16", "char_start": 4002, "verified_on": "2026-11-13", "observed": "page", "source_work": { "title": "plinder-org/runs-n-poses README", "authors_or_org": "Škrinjar P, Eberhardt J, Tauriello G, Schwede T, Durairaj J (PLINDER org)", "date": null, "date_as_written": null, "source_type": "official-page", "open_access": true, "doi_or_url": "https://github.com/plinder-org/runs-n-poses" } } ] } } } ], "access": [ { "method": "open-weights-download", "url": "https://storage.googleapis.com/alphafold3/af3.bin.zst", "terms_differ_from_weights_license": false, "own_terms": null }, { "method": "hosted-api", "url": "https://alphafoldserver.com", "terms_differ_from_weights_license": false, "own_terms": null }, { "method": "third-party-hosted", "url": "https://github.com/google-deepmind/alphafold3/blob/main/README.md", "terms_differ_from_weights_license": false, "own_terms": { "text": { "en": "AlphaFold Server (alphafoldserver.com) offers AF3 for non-commercial use with a more limited set of ligands and modifications; commercial use is offered through Google Cloud.", "zh": "AlphaFold Server（alphafoldserver.com）提供非商业用途的 AF3，配体与修饰种类较少；商业使用经由 Google Cloud 提供。" }, "evidence": [ { "location": "dataset-page", "excerpt": "for non-commercial use, though with a more limited set of ligands and covalent modifications.", "url": "https://github.com/google-deepmind/alphafold3/blob/main/README.md", "snapshot_id": "s10", "char_start": null, "verified_on": "2026-11-13", "observed": "page", "source_work": { "title": "AlphaFold 3 README", "authors_or_org": "Google DeepMind", "date": "2026-08-21", "date_as_written": "2026-08-21 (Cloud links added)", "source_type": "official-page", "open_access": true, "doi_or_url": "https://github.com/google-deepmind/alphafold3/blob/main/README.md" } } ] } } ], "conflicts": [], "hardware_notes": [ { "text": { "en": "Requires Linux, up to 1 TB disk for genetic databases and an NVIDIA GPU with Compute Capability 8.0+; inputs up to 5,120 tokens verified on a single A100 80 GB or H100 80 GB.", "zh": "需要 Linux、最多 1 TB 磁盘存放遗传数据库、计算能力 8.0 及以上的 NVIDIA GPU；已验证单张 A100 80GB 或 H100 80GB 可处理最多 5,120 个 token 的输入。" }, "evidence": [ { "location": "dataset-page", "excerpt": "We have verified that inputs with up to 5,120 tokens can fit on a single NVIDIA A100 80 GB, or a single NVIDIA H100 80 GB.", "url": "https://github.com/google-deepmind/alphafold3/blob/main/docs/installation.md", "snapshot_id": "s17", "char_start": null, "verified_on": "2026-11-13", "observed": "page", "source_work": { "title": "AlphaFold 3 installation documentation", "authors_or_org": "Google DeepMind", "date": null, "date_as_written": null, "source_type": "official-page", "open_access": true, "doi_or_url": "https://github.com/google-deepmind/alphafold3/blob/main/docs/installation.md" } } ] } ], "known_defects": [], "fit": { "suited_for": [ { "text": { "en": "Proteins with shallow MSAs are predicted with lower accuracy (AF3 paper).", "zh": "MSA 较浅的蛋白预测精度较低（AF3 论文）。" }, "evidence": [ { "location": "peer-reviewed-paper", "excerpt": "proteins with shallow MSAs are predicted with lower accuracy", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC11168924/", "snapshot_id": "s11", "char_start": 39846, "verified_on": "2026-11-13", "observed": "page", "source_work": { "title": "Accurate structure prediction of biomolecular interactions with AlphaFold 3", "authors_or_org": "Abramson J, Adler J, Dunger J, et al. (Nature)", "date": "2024-05-08", "date_as_written": "2024-05-08", "source_type": "peer-reviewed-paper", "open_access": true, "doi_or_url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC11168924/" } } ] } ], "not_suited_for": [] }, "gaps": [ { "field": "known_defects", "note": "未找到有出处的说明 / no sourced statement found" } ], "completeness": { "core_complete": true }, "changelog": [ { "date": "2026-11-14", "change": { "en": "Listed on 2026-11-14. Judged by public rules and checked a second time before listing. The two passes disagreed; the stricter reading is carried.", "zh": "2026-11-14 上架。按公开规则判定，经二次核对后收录。两遍判定不一致，卡上取较严的读法。" }, "by": "second-pass" } ], "license_review": { "method": "rules-plus-second-pass-model", "method_label": { "en": "Judged by public rules and checked a second time before listing.", "zh": "按公开规则判定，经二次核对后收录。" }, "readings": [ { "restatement": { "en": "First pass (rules).", "zh": "第一遍（规则）。" }, "quotes": [ { "excerpt": "The AlphaFold 3 model parameters and output are **only** available for", "snapshot_id": "s12", "url": "https://raw.githubusercontent.com/google-deepmind/alphafold3/main/WEIGHTS_TERMS_OF_USE.md" }, { "excerpt": "You ***must not* publish or share AlphaFold 3 model parameters**", "snapshot_id": "s12", "url": "https://raw.githubusercontent.com/google-deepmind/alphafold3/main/WEIGHTS_TERMS_OF_USE.md" }, { "excerpt": "The AlphaFold 3 model parameters and output are **only** available for non-commercial use by, or on behalf of, non-commercial organizations (*i.e.*, universities, non-profit organizations and research institutes, educational, journalism and government bodies).", "snapshot_id": "s08", "url": "https://raw.githubusercontent.com/google-deepmind/alphafold3/main/WEIGHTS_TERMS_OF_USE.md" } ], "values": { "license_status": "restricted", "commercial_use": "no", "derivatives": "unstated", "redistribution": "unstated" }, "source": "rules" }, { "restatement": { "en": "Second pass: weights.redistribution: 原文明确不得发布或分享模型参数（仅可在本组织内按条款分享），个人获授权者不得与任何人分享。第一遍把该句列为证据却填 unstated。应为 no。 | missed sentence (class A): You must not share Output with any commercial organization. | missed sentence (class D): Distribute Output, or disclose findings arising from using AlphaFold 3 | missed sentence (class A?): making decisions in domains that affect material or individual rights | missed sentence (class D): not (i) resident of a embargoed country, (ii) ordinarily resident in a | missed sentence (class A?): Google may at any time suspend or terminate your right to use and, as | missed sentence (class A): You may only use AlphaFold 3 model parameters if received directly fro", "zh": "第二遍复核：weights.redistribution: 原文明确不得发布或分享模型参数（仅可在本组织内按条款分享），个人获授权者不得与任何人分享。第一遍把该句列为证据却填 unstated。应为 no。；漏读句（A 类）：You must not share Output with any commercial organization.；漏读句（D 类）：Distribute Output, or disclose findings arising from using AlphaFold 3；漏读句（A? 类）：making decisions in domains that affect material or individual rights ；漏读句（D 类）：not (i) resident of a embargoed country, (ii) ordinarily resident in a；漏读句（A? 类）：Google may at any time suspend or terminate your right to use and, as ；漏读句（A 类）：You may only use AlphaFold 3 model parameters if received directly fro" }, "quotes": [ { "excerpt": "You must not share Output with any commercial organization.", "snapshot_id": "s15", "url": null }, { "excerpt": "Distribute Output, or disclose findings arising from using AlphaFold 3 without citing our paper:", "snapshot_id": "s05", "url": null }, { "excerpt": "making decisions in domains that affect material or individual rights or", "snapshot_id": "s15", "url": null }, { "excerpt": "not (i) resident of a embargoed country, (ii) ordinarily resident in a US", "snapshot_id": "s08", "url": null } ], "values": { "license_status": "restricted", "commercial_use": "no", "derivatives": "unstated", "redistribution": "no" }, "source": "second-pass" } ], "adopted": 1, "second_pass": { "reviewer": "model-second-pass", "reviewed_at": "2026-11-13", "overall": "stricter", "agreed": false, "disagreement": { "en": "Second pass: weights.redistribution: 原文明确不得发布或分享模型参数（仅可在本组织内按条款分享），个人获授权者不得与任何人分享。第一遍把该句列为证据却填 unstated。应为 no。 | missed sentence (class A): You must not share Output with any commercial organization. | missed sentence (class D): Distribute Output, or disclose findings arising from using AlphaFold 3 | missed sentence (class A?): making decisions in domains that affect material or individual rights | missed sentence (class D): not (i) resident of a embargoed country, (ii) ordinarily resident in a | missed sentence (class A?): Google may at any time suspend or terminate your right to use and, as | missed sentence (class A): You may only use AlphaFold 3 model parameters if received directly fro", "zh": "第二遍复核：weights.redistribution: 原文明确不得发布或分享模型参数（仅可在本组织内按条款分享），个人获授权者不得与任何人分享。第一遍把该句列为证据却填 unstated。应为 no。；漏读句（A 类）：You must not share Output with any commercial organization.；漏读句（D 类）：Distribute Output, or disclose findings arising from using AlphaFold 3；漏读句（A? 类）：making decisions in domains that affect material or individual rights ；漏读句（D 类）：not (i) resident of a embargoed country, (ii) ordinarily resident in a；漏读句（A? 类）：Google may at any time suspend or terminate your right to use and, as ；漏读句（A 类）：You may only use AlphaFold 3 model parameters if received directly fro" }, "missed_restrictions": [ { "excerpt": "You must not share Output with any commercial organization.", "snapshot_id": "s15", "class": "A", "kind": "restriction" }, { "excerpt": "Distribute Output, or disclose findings arising from using AlphaFold 3 without citing our paper:", "snapshot_id": "s05", "class": "D", "kind": "noise" }, { "excerpt": "making decisions in domains that affect material or individual rights or", "snapshot_id": "s15", "class": "A?", "kind": "unclear" }, { "excerpt": "not (i) resident of a embargoed country, (ii) ordinarily resident in a US", "snapshot_id": "s08", "class": "D", "kind": "noise" }, { "excerpt": "Google may at any time suspend or terminate your right to use and, as applicable", "snapshot_id": "s08", "class": "A?", "kind": "unclear" }, { "excerpt": "You may only use AlphaFold 3 model parameters if received directly from Google.", "snapshot_id": "s02", "class": "A", "kind": "restriction" } ] } }, "access_recipe": { "verified_on": "2026-11-14", "platform": "github", "repo_id": "google-deepmind/alphafold3", "source": { "api_url": "https://api.github.com/repos/google-deepmind/alphafold3", "fetched_on": "2026-11-14" }, "access_state": "application-required", "files": [], "files_truncated": false, "commands": [ { "command": "git clone https://github.com/google-deepmind/alphafold3", "evidence": { "excerpt": "git clone https://github.com/YOUR-USERNAME/YOUR-REPOSITORY", "doc": "github-clone", "url": "https://docs.github.com/en/repositories/creating-and-managing-repositories/cloning-a-repository", "fetched_on": "2026-11-14" } } ], "registration": { "required": true, "url": "https://forms.gle/svvpY4u2jsHEwWYS6", "steps": [ "The model parameters are not downloadable freely: a request form has to be completed and access is granted by the owner (we give no way around this)" ], "evidence": { "excerpt": "To request access to the AlphaFold 3 model parameters, please complete", "snapshot_id": "s14", "url": "https://github.com/google-deepmind/alphafold3/commit/dd1a7badb6" } }, "version": { "commit": null, "last_modified": "2026-11-13T15:55:23Z", "default_branch": "main" }, "latest_release": { "version": "v3.0.5", "date": "2026-10-30T15:31:31Z" }, "license_note": { "license_status": "restricted", "text": "Use according to the terms of the official source; this recipe does not say the data can be used freely." }, "gaps": [ "the platform publishes no checksum for these files; none was computed (we do not download data files)" ] }, "_links": { "text": "Continue in this database. This record is one of 109 processed and verified records (datasets, data resources and open models). To find others, call `https://mcp.mdexa.com/v1/datasets?q=` or connect the MCP server at `https://mcp.mdexa.com/mcp`; no key is needed for 20 calls per minute. Rows are under 1 KB; an empty result means \"not listed\". For a list of items, one batch check (free account) returns a report. Guide: /agents.md. Similar records: Boltz (Boltz-1 and Boltz-2) (https://mdexa.com/models/boltz/); OpenFold3 (https://mdexa.com/models/openfold3/); ESM C (https://mdexa.com/models/esm-c/).", "records": 109, "search": { "template": "https://mcp.mdexa.com/v1/datasets?q={words}", "example": "https://mcp.mdexa.com/v1/datasets?q=binding%20affinity" }, "mcp": "https://mcp.mdexa.com/mcp", "anonymous_calls_per_minute": 20, "guide": "https://mdexa.com/agents.md", "faq": "https://mdexa.com/developers/faq/", "similar": [ { "name": "Boltz (Boltz-1 and Boltz-2)", "url": "https://mdexa.com/models/boltz/", "json": "https://mdexa.com/models/boltz.json" }, { "name": "OpenFold3", "url": "https://mdexa.com/models/openfold3/", "json": "https://mdexa.com/models/openfold3.json" }, { "name": "ESM C", "url": "https://mdexa.com/models/esm-c/", "json": "https://mdexa.com/models/esm-c.json" } ] } }