{"@context":"https://schema.org","@type":"NewsArticle","generatedAt":"2026-07-28T06:20:51.496Z","headline":"FAIRChem v2 UMA 通用机器学习原子间势发布","description":"FAIRChem v2 推出 UMA 通用机器学习原子间势，可跨分子化学、催化和无机材料三大领域执行单点能量预测、几何优化、分子动力学等任务。该框架通过 Hugging Face 认证获取 gated 模型权重，并集成 Atomic Simulation Environment 管理原子结构与轨迹分析。","url":"https://www.aioga.com/news/cms1ke8td00c6ro05i2sb5cn4/","mainEntityOfPage":"https://www.aioga.com/news/cms1ke8td00c6ro05i2sb5cn4/","datePublished":"2026-07-26T08:38:09.000Z","dateModified":"2026-07-26T08:38:09.000Z","inLanguage":"zh-CN","publisher":{"@type":"NewsMediaOrganization","name":"Aioga","url":"https://www.aioga.com"},"citation":["https://www.marktechpost.com/2026/07/26/fairchem-v2-uma-for-multidomain-atomistic-simulation-across-molecules-catalysts-materials-vibrations-and-molecular-dynamics","https://aihot.virxact.com/items/cms1ke8td00c6ro05i2sb5cn4"],"canonicalUrl":"https://www.aioga.com/news/cms1ke8td00c6ro05i2sb5cn4/","directAnswer":{"@type":"Answer","text":"Aioga 编辑摘要：FAIRChem v2 推出 UMA 通用机器学习原子间势，可跨分子化学、催化和无机材料三大领域执行单点能量预测、几何优化、分子动力学等任务。 Aioga 将其归入「技巧观点」方向，重点关注它对真实使用和行业竞争的影响。","url":"https://www.aioga.com/news/cms1ke8td00c6ro05i2sb5cn4/","dateCreated":"2026-07-26T08:38:09.000Z","author":{"@type":"Organization","@id":"https://www.aioga.com/authors/aioga-editorial/#editorial-team","name":"Aioga Editorial Team","url":"https://www.aioga.com/authors/aioga-editorial/"}},"evidence":[{"@type":"CreativeWork","name":"marktechpost.com source article","url":"https://www.marktechpost.com/2026/07/26/fairchem-v2-uma-for-multidomain-atomistic-simulation-across-molecules-catalysts-materials-vibrations-and-molecular-dynamics","datePublished":"2026-07-26T08:38:09.000Z","provider":{"@type":"Organization","name":"marktechpost.com","url":"https://www.marktechpost.com/2026/07/26/fairchem-v2-uma-for-multidomain-atomistic-simulation-across-molecules-catalysts-materials-vibrations-and-molecular-dynamics"}},{"@type":"CreativeWork","name":"AIHot archive record","url":"https://aihot.virxact.com/items/cms1ke8td00c6ro05i2sb5cn4","datePublished":"2026-07-26T08:38:09.000Z","provider":{"@type":"Organization","name":"AIHot","url":"https://aihot.virxact.com/items/cms1ke8td00c6ro05i2sb5cn4"}}],"aggregationSource":"MarkTechPost（RSS）","originalPublisher":{"name":"marktechpost.com","url":"https://www.marktechpost.com/2026/07/26/fairchem-v2-uma-for-multidomain-atomistic-simulation-across-molecules-catalysts-materials-vibrations-and-molecular-dynamics"},"article":{"id":"cms1ke8td00c6ro05i2sb5cn4","slug":"cms1ke8td00c6ro05i2sb5cn4","url":"https://www.aioga.com/news/cms1ke8td00c6ro05i2sb5cn4/","title":"FAIRChem v2 UMA 通用机器学习原子间势发布","title_en":"FAIRChem v2 UMA for Multidomain Atomistic Simulation across Molecules， Catalysts， Materials， Vibrations， and Molecular Dynamics","summary":"FAIRChem v2 推出 UMA 通用机器学习原子间势，可跨分子化学、催化和无机材料三大领域执行单点能量预测、几何优化、分子动力学等任务。该框架通过 Hugging Face 认证获取 gated 模型权重，并集成 Atomic Simulation Environment 管理原子结构与轨迹分析。","source":"MarkTechPost（RSS）","sourceUrl":"https://www.marktechpost.com/2026/07/26/fairchem-v2-uma-for-multidomain-atomistic-simulation-across-molecules-catalysts-materials-vibrations-and-molecular-dynamics","aiHotUrl":"https://aihot.virxact.com/items/cms1ke8td00c6ro05i2sb5cn4","publishedAt":"2026-07-26T08:38:09.000Z","category":"技巧观点","score":59,"selected":false,"articleBody":["In this tutorial, we explore FAIRChem v2 ：https://github.com/facebookresearch/fairchem and the UMA universal machine-learning interatomic potential as a unified framework for atomistic simulation across molecular chemistry, catalysis, and inorganic materials. We configure an environment, authenticate with Hugging Face to access the gated UMA model weights, and initialize task-specific calculators for the omol, oc20, and omat domains. We then apply the same pretrained potential to a broad set of computational chemistry workflows, including single-point energy and force prediction, molecular geometry optimization, spin-state comparison, reaction-energy estimation, vibrational analysis, surface adsorption, crystal-cell relaxation, equation-of-state fitting, molecular dynamics, and potential-energy surface scanning. Throughout the tutorial, we integrate FAIRChem with the Atomic Simulation Environment to manage atomic structures, optimizers, constraints, thermodynamic calculations, and trajectory analysis while using GPU acceleration whenever it is available.","We install the required FAIRChem, ASE, visualization, and Hugging Face dependencies while ensuring the setup remains safe to rerun in Google Colab. We authenticate with Hugging Face to access the gated UMA model weights and automatically detect whether GPU acceleration is available. We then load the UMA predictor and create separate calculators for molecular, catalysis, and materials simulations.","We use the molecular UMA calculator to evaluate the energy, atomic forces, and atomization energy of a water molecule. We define isolated hydrogen and oxygen reference atoms with the correct spin multiplicities to construct the atomization-energy calculation. We then distort the water geometry, relax it with the LBFGS optimizer, and analyze the converged bond length, bond angle, and energy trajectory.","We compare the singlet and triplet electronic states of methylene to calculate its spin-state energy gap. We relax methane, oxygen, carbon dioxide, and water before combining their predicted energies to estimate the electronic reaction energy of methane combustion. We also perform a finite-difference vibrational analysis of relaxed water to obtain its normal-mode frequencies and zero-point energy.","We construct a periodic Cu(100) slab, place a carbon monoxide molecule at a bridge adsorption site, and constrain the lower copper layers. We relax the adsorbate–surface system with the OC20 calculator and separately optimize the clean slab and gas-phase carbon monoxide references. We then evaluate a pedagogical adsorption-energy cycle while recognizing that OC20 uses a task-specific energy reference convention.","We relax the atomic positions and simulation cell of BCC iron using the materials-domain UMA calculator and a Frechet cell filter. We sample energies across a range of compressed and expanded volumes, fit a Birch–Murnaghan equation of state, and estimate the equilibrium volume and bulk modulus. We also run Langevin molecular dynamics for water at 300 K and track its temperature, potential energy, and O–H bond length over time.","We scan the potential-energy surface of water by systematically stretching one O–H bond across a selected distance range. We evaluate the molecular energy at each geometry, normalize the energies relative to the minimum, and visualize the resulting dissociation profile. We conclude the tutorial by identifying higher-accuracy UMA models, additional chemical domains, and larger multi-GPU simulations as possible extensions.","In conclusion, we built a complete atomistic simulation workflow around FAIRChem v2 and demonstrated how UMA provides a shared learned potential across chemically distinct domains without requiring a separate model for every task. We used molecular calculations to evaluate energies, forces, atomization behavior, spin gaps, reaction energetics, vibrational modes, and bond-stretch profiles; we used the catalysis domain to relax CO on a Cu(100) surface and examine adsorption energetics; and we used the materials domain to relax BCC iron and estimate its bulk modulus from an equation-of-state fit. We also ran Langevin molecular dynamics to observe finite-temperature structural and energetic fluctuations over time. By combining UMA inference with ASE structure builders, optimizers, filters, vibrational tools, and molecular-dynamics utilities, we established a reusable foundation for extending the workflow to larger molecules, catalytic interfaces, crystalline materials, metal-organic frameworks, molecular crystals, and higher-accuracy UMA model variants.","Need to partner with us for promoting your GitHub Repo OR Hugging Face Page OR Product Release OR Webinar etc.? Connect with us ：https://forms.gle/wbash1wF6efRj8G58","Sana Hassan, a consulting intern at Marktechpost and dual-degree student at IIT Madras, is passionate about applying technology and AI to address real-world challenges. With a keen interest in solving practical problems, he brings a fresh perspective to the intersection of AI and real-life solutions.","Build an Agentic Event Venue Operator [Full Codes]：https://pxllnk.co/twdn5","Thanks! Our team will contact you soon 🙌"],"articleImages":[{"sourceUrl":"https://www.marktechpost.com/wp-content/uploads/2026/07/blog6171-1-100x70.png","alt":"Designing High-Performance GPU Kernels with TileLang: Tensor-Core GEMM, Fused Softmax, FlashAttention, and Autotuning","afterParagraph":9,"url":"/media/articles/cms1ke8td00c6ro05i2sb5cn4/80925b48ffa65ad2.webp"}],"mediaStatus":"ok","articleBodyZh":["在本教程中，我们探索 FAIRChem v2：https://github.com/facebookresearch/fairchem 以及 UMA 通用机器学习原子间势作为一个统一的原子模拟框架，涵盖分子化学、催化和无机材料。我们配置环境，使用 Hugging Face 进行身份验证以访问受限的 UMA 模型权重，并为 omol、oc20 和 omat 领域初始化特定任务计算器。然后，我们将相同的预训练势应用于广泛的计算化学工作流程，包括单点能量和力预测、分子几何优化、自旋态比较、反应能估算、振动分析、表面吸附、晶胞弛豫、状态方程拟合、分子动力学以及势能面扫描。在整个教程中，我们将 FAIRChem 与原子模拟环境（Atomic Simulation Environment, ASE）集成，用于管理原子结构、优化器、约束、热力学计算和轨迹分析，同时在可用时使用 GPU 加速。","我们安装所需的 FAIRChem、ASE、可视化和 Hugging Face 依赖项，同时确保在 Google Colab 中重复运行时环境安全。我们使用 Hugging Face 进行身份验证以访问受限的 UMA 模型权重，并自动检测 GPU 加速是否可用。然后加载 UMA 预测器，并为分子、催化和材料模拟创建独立的计算器。","我们使用分子 UMA 计算器评估水分子的能量、原子力和成键能。我们定义具有正确自旋重数的孤立氢和氧参考原子，以构建成键能计算。随后，我们扭曲水分子的几何结构，用 LBFGS 优化器进行弛豫，并分析收敛后的键长、键角和能量轨迹。","我们比较亚甲基的单重态和三重态电子状态，以计算其自旋态能量差。我们对甲烷、氧气、二氧化碳和水进行弛豫，然后将它们的预测能量组合以估算甲烷燃烧的电子反应能。同时，我们还对弛豫后的水进行有限差分振动分析，以获得其正常模式频率和零点能。","我们构建了一个周期性Cu(100)晶片，将一氧化碳分子放置在桥位吸附位点，并约束下层铜原子层。我们使用OC20计算器放松吸附物–表面系统，单独优化干净晶片和气态一氧化碳的参考体系。然后我们评估了一个教学性的吸附能量循环，同时注意到OC20使用了任务特定的能量参考约定。","我们使用材料领域的UMA计算器和Frechet单元滤波器放松BCC铁的原子位置和模拟单元。我们在不同压缩和膨胀体积范围内采样能量，拟合Birch–Murnaghan方程，估计平衡体积和体积模量。我们还对300 K下的水进行了Langevin分子动力学模拟，并随时间追踪其温度、势能和O–H键长。","我们通过系统地拉伸一个O–H键在选定的距离范围内扫描水的势能面。我们在每个几何结构下评估分子能量，相对于最小值对能量进行归一化，并可视化得到的解离曲线。我们通过确认更高精度的UMA模型、额外的化学领域以及更大规模的多GPU模拟，来结束本教程作为可能的扩展。","总之，我们围绕FAIRChem v2构建了完整的原子模拟工作流程，并展示了UMA如何在不同化学领域提供共享的学习势，而无需为每项任务单独建立模型。我们使用分子计算评估能量、力、原子化行为、自旋间隙、反应能学、振动模式和键拉伸曲线；使用催化领域放松Cu(100)表面上的CO并检查吸附能；使用材料领域放松BCC铁并通过方程拟合估计其体积模量。我们还进行了Langevin分子动力学模拟，以观察有限温度下的结构和能量波动。通过将UMA推理与ASE结构构建器、优化器、滤波器、振动工具和分子动力学工具结合，我们建立了一个可重复使用的基础，用于将工作流程扩展到更大分子、催化界面、晶体材料、金属有机框架、分子晶体以及更高精度的UMA模型变体。","需要与我们合作以推广您的 GitHub 仓库或 Hugging Face 页面或产品发布或网络研讨会等吗？请联系：https://forms.gle/wbash1wF6efRj8G58","Sana Hassan 是 Marktechpost 的咨询实习生，同时也是 IIT Madras 的双学位学生，他热衷于将技术和人工智能应用于解决现实世界的挑战。凭借对解决实际问题的浓厚兴趣，他为人工智能与现实生活解决方案的交汇点带来了新的视角。","构建一个智能事件场地运营者 [完整代码]：https://pxllnk.co/twdn5","谢谢！我们的团队将很快与您联系 🙌"],"translationStatus":"translated","bodyOrigin":"source-page","editorial":{"summary":"Aioga 编辑摘要：FAIRChem v2 推出 UMA 通用机器学习原子间势，可跨分子化学、催化和无机材料三大领域执行单点能量预测、几何优化、分子动力学等任务。 Aioga 将其归入「技巧观点」方向，重点关注它对真实使用和行业竞争的影响。","background":"背景分析：实践类内容的价值在于是否能被复现、是否有明确边界，以及它能否转化为稳定的开发或工作流方法。","viewpoint":"Aioga 判断：这条动态更适合作为行业观察信号，当前信息足以建立线索，但不足以推导长期结论。","implications":"影响分析：对相关团队而言，短期应先核对来源、可用范围和实际成本，再判断是否值得接入或跟进。","nextStep":"后续观察：继续观察示例是否可复现、工具版本变化、社区反馈和实际成本。","evidenceRefs":["title","summary","articleBody"],"confidence":"medium","status":"published","aiGenerated":false,"autoApproved":true,"generatedBy":"rule-safe-fallback","generatedAt":"2026-07-28T06:29:10.489Z","sourceHash":"3ad5ffb7929cd503","validation":{"passed":true,"mode":"rule-safe-fallback","checks":["schema","length","source-attribution","no-html"]}},"tags":["技巧观点","MarkTechPost（RSS）"],"translations":{"zh-CN":{"title":"FAIRChem v2 UMA 通用机器学习原子间势发布","summary":"FAIRChem v2 推出 UMA 通用机器学习原子间势，可跨分子化学、催化和无机材料三大领域执行单点能量预测、几何优化、分子动力学等任务。该框架通过 Hugging Face 认证获取 gated 模型权重，并集成 Atomic Simulation Environment 管理原子结构与轨迹分析。","category":"技巧观点","source":"marktechpost.com","aggregationSource":"MarkTechPost（RSS）","pageTitle":"FAIRChem v2 UMA 通用机器学习原子间势发布 - Aioga AI资讯","description":"FAIRChem v2 推出 UMA 通用机器学习原子间势，可跨分子化学、催化和无机材料三大领域执行单点能量预测、几何优化、分子动力学等任务。该框架通过 Hugging Face 认证获取 gated 模型权重，并集成 Atomic Simulation Environment 管理原子结构与轨迹分析。","url":"https://www.aioga.com/news/cms1ke8td00c6ro05i2sb5cn4/"},"en":{"title":"FAIRChem v2 UMA General Machine Learning Interatomic Potential Released","summary":"FAIRChem v2 launches the UMA universal machine learning interatomic potential, capable of performing single-point energy predictions, geometry optimization, molecular dynamics, and other tasks across the three major fields of molecular chemistry, catalysis, and inorganic materials. This framework obtains gated model weights through Hugging Face certification and integrates Atomic Simulation Environment to manage atomic structures and trajectory analysis.","category":"Insights","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"FAIRChem v2 UMA General Machine Learning Interatomic Potential Released - Aioga AI News","description":"FAIRChem v2 launches the UMA universal machine learning interatomic potential, capable of performing single-point energy predictions, geometry optimization, molecular dynamics, and...","url":"https://www.aioga.com/en/news/cms1ke8td00c6ro05i2sb5cn4/","contentTranslated":true,"sourceHash":"b4ce96e5c2241881","translatedAt":"2026-07-26T10:22:00.357Z"},"ja":{"title":"FAIRChem v2 UMA 汎用機械学習原子間ポテンシャル公開","summary":"FAIRChem v2 は UMA 汎用機械学習原子間ポテンシャルを導入し、分子化学、触媒、無機材料の三大分野で単一点エネルギー予測、幾何最適化、分子動力学などのタスクを実行できます。このフレームワークは Hugging Face 認証を通じてゲート付きモデルの重みを取得し、Atomic Simulation Environment を統合して原子構造と軌道解析を管理します。","category":"ヒントと視点","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"FAIRChem v2 UMA 汎用機械学習原子間ポテンシャル公開 - Aioga AIニュース","description":"FAIRChem v2 は UMA 汎用機械学習原子間ポテンシャルを導入し、分子化学、触媒、無機材料の三大分野で単一点エネルギー予測、幾何最適化、分子動力学などのタスクを実行できます。このフレームワークは Hugging Face 認証を通じてゲート付きモデルの重みを取得し、Atomic Simulation Environment を統合して原子構造と軌道...","url":"https://www.aioga.com/ja/news/cms1ke8td00c6ro05i2sb5cn4/","contentTranslated":true,"sourceHash":"b4ce96e5c2241881","translatedAt":"2026-07-26T10:22:12.365Z"},"ko":{"title":"FAIRChem v2 UMA 범용 머신러닝 원자간 포텐셜 발표","summary":"FAIRChem v2는 UMA 범용 기계 학습 원자 간 포텐셜을 출시하여 분자 화학, 촉매, 무기 재료 세 가지 주요 분야에서 단일 포인트 에너지 예측, 기하학 최적화, 분자 역학 등의 작업을 수행할 수 있습니다. 이 프레임워크는 Hugging Face 인증을 통해 gated 모델 가중치를 얻고, Atomic Simulation Environment을 통합하여 원자 구조와 궤적 분석을 관리합니다.","category":"인사이트","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"FAIRChem v2 UMA 범용 머신러닝 원자간 포텐셜 발표 - Aioga AI 뉴스","description":"FAIRChem v2는 UMA 범용 기계 학습 원자 간 포텐셜을 출시하여 분자 화학, 촉매, 무기 재료 세 가지 주요 분야에서 단일 포인트 에너지 예측, 기하학 최적화, 분자 역학 등의 작업을 수행할 수 있습니다. 이 프레임워크는 Hugging Face 인증을 통해 gated 모델 가중치를 얻고, Atomic Simula...","url":"https://www.aioga.com/ko/news/cms1ke8td00c6ro05i2sb5cn4/","contentTranslated":true,"sourceHash":"b4ce96e5c2241881","translatedAt":"2026-07-26T10:22:54.119Z"},"es":{"title":"FAIRChem v2 UMA Lanzamiento de potenciales interatómicos de aprendizaje automático general","summary":"FAIRChem v2 ha lanzado el potencial atómico de aprendizaje automático general UMA, que puede realizar tareas como predicción de energía puntual, optimización geométrica y dinámica molecular en los tres grandes campos de la química molecular, la catálisis y los materiales inorgánicos. Este marco obtiene los pesos del modelo con acceso restringido a través de la certificación de Hugging Face, e integra Atomic Simulation Environment para gestionar estructuras atómicas y análisis de trayectorias.","category":"Ideas","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"FAIRChem v2 UMA Lanzamiento de potenciales interatómicos de aprendizaje automático general - Aioga Noticias de IA","description":"FAIRChem v2 ha lanzado el potencial atómico de aprendizaje automático general UMA, que puede realizar tareas como predicción de energía puntual, optimización geométrica y dinámica...","url":"https://www.aioga.com/es/news/cms1ke8td00c6ro05i2sb5cn4/","contentTranslated":true,"sourceHash":"b4ce96e5c2241881","translatedAt":"2026-07-26T10:22:49.065Z"},"fr":{"title":"Publication du potentiel atomique d'apprentissage automatique général FAIRChem v2 UMA","summary":"FAIRChem v2 a lancé le potentiel interatomique d'apprentissage automatique universel UMA, capable d'exécuter des prédictions d'énergie ponctuelle, l'optimisation géométrique, la dynamique moléculaire et d'autres tâches dans les trois grands domaines de la chimie moléculaire, de la catalyse et des matériaux inorganiques. Ce cadre obtient les poids du modèle gated via la certification Hugging Face et intègre Atomic Simulation Environment pour gérer les structures atomiques et l'analyse des trajectoires.","category":"Analyses","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"Publication du potentiel atomique d'apprentissage automatique général FAIRChem v2 UMA - Aioga Actualités IA","description":"FAIRChem v2 a lancé le potentiel interatomique d'apprentissage automatique universel UMA, capable d'exécuter des prédictions d'énergie ponctuelle, l'optimisation géométrique, la dy...","url":"https://www.aioga.com/fr/news/cms1ke8td00c6ro05i2sb5cn4/","contentTranslated":true,"sourceHash":"b4ce96e5c2241881","translatedAt":"2026-07-26T10:23:34.108Z"},"de":{"title":"FAIRChem v2 UMA Allgemeine maschinelle Lernpotenziale zwischen Atomen veröffentlicht","summary":"FAIRChem v2 hat das UMA Universal Machine Learning Interatomic Potential eingeführt, das energiespezifische Vorhersagen, Geometrieoptimierungen, Molekulardynamik und andere Aufgaben über die drei Hauptbereiche Molekülchemie, Katalyse und anorganische Materialien hinweg ausführen kann. Dieses Framework bezieht die freigeschalteten Modellgewichte über die Zertifizierung von Hugging Face und integriert die Atomic Simulation Environment zur Verwaltung von Atomstrukturen und Trajektorienanalysen.","category":"技巧观点","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"FAIRChem v2 UMA Allgemeine maschinelle Lernpotenziale zwischen Atomen veröffentlicht - Aioga KI-News","description":"FAIRChem v2 hat das UMA Universal Machine Learning Interatomic Potential eingeführt, das energiespezifische Vorhersagen, Geometrieoptimierungen, Molekulardynamik und andere Aufgabe...","url":"https://www.aioga.com/de/news/cms1ke8td00c6ro05i2sb5cn4/","contentTranslated":true,"sourceHash":"b4ce96e5c2241881","translatedAt":"2026-07-26T10:23:28.928Z"},"pt-BR":{"title":"Lançamento do potencial interatômico de aprendizado de máquina geral FAIRChem v2 UMA","summary":"FAIRChem v2 lançou o potencial atômico de aprendizado de máquina universal UMA, capaz de executar previsões de energia pontual, otimização geométrica, dinâmica molecular e outras tarefas nos três principais campos de química molecular, catálise e materiais inorgânicos. A estrutura obtém pesos de modelo gated certificados pelo Hugging Face e integra o Atomic Simulation Environment para gerenciar estruturas atômicas e análise de trajetórias.","category":"技巧观点","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"Lançamento do potencial interatômico de aprendizado de máquina geral FAIRChem v2 UMA - Aioga Notícias de IA","description":"FAIRChem v2 lançou o potencial atômico de aprendizado de máquina universal UMA, capaz de executar previsões de energia pontual, otimização geométrica, dinâmica molecular e outras t...","url":"https://www.aioga.com/pt-BR/news/cms1ke8td00c6ro05i2sb5cn4/","contentTranslated":true,"sourceHash":"b4ce96e5c2241881","translatedAt":"2026-07-26T10:24:11.465Z"},"ru":{"title":"Выпуск универсального машинного обучения межатомных потенциалов FAIRChem v2 UMA","summary":"FAIRChem v2 представляет UMA универсальный машинно-обучаемый потенциал межатомных взаимодействий, который может выполнять задачи одноточечной оценки энергии, геометрической оптимизации, молекулярной динамики и другие в трех основных областях: молекулярная химия, катализм и неорганические материалы. Эта платформа получает веса моделей с ограниченным доступом через сертификацию Hugging Face и интегрирует Atomic Simulation Environment для управления атомными структурами и анализа траекторий.","category":"技巧观点","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"Выпуск универсального машинного обучения межатомных потенциалов FAIRChem v2 UMA - Aioga Новости ИИ","description":"FAIRChem v2 представляет UMA универсальный машинно-обучаемый потенциал межатомных взаимодействий, который может выполнять задачи одноточечной оценки энергии, геометрической оптимиз...","url":"https://www.aioga.com/ru/news/cms1ke8td00c6ro05i2sb5cn4/","contentTranslated":true,"sourceHash":"b4ce96e5c2241881","translatedAt":"2026-07-26T10:24:14.062Z"},"ar":{"title":"إصدار FAIRChem v2 UMA لإمكانات بين الذرات باستخدام التعلم الآلي العام","summary":"أطلق FAIRChem v2 إمكانات الذكاء الاصطناعي العام للتعلم الآلي بين الذرات UMA، والتي يمكنها تنفيذ التنبؤ بالطاقة النقطة الواحدة، والتحسين الهندسي، والديناميكيات الجزيئية عبر ثلاثة مجالات رئيسية: كيمياء الجزيئات، والتحفيز، والمواد غير العضوية. يقوم هذا الإطار بالحصول على أوزان النماذج gated من خلال اعتماد Hugging Face، ويقوم بتكامل بيئة المحاكاة الذرية لإدارة الهياكل الذرية وتحليل المسارات.","category":"技巧观点","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"إصدار FAIRChem v2 UMA لإمكانات بين الذرات باستخدام التعلم الآلي العام - Aioga أخبار الذكاء الاصطناعي","description":"أطلق FAIRChem v2 إمكانات الذكاء الاصطناعي العام للتعلم الآلي بين الذرات UMA، والتي يمكنها تنفيذ التنبؤ بالطاقة النقطة الواحدة، والتحسين الهندسي، والديناميكيات الجزيئية عبر ثلاثة مج...","url":"https://www.aioga.com/ar/news/cms1ke8td00c6ro05i2sb5cn4/","contentTranslated":true,"sourceHash":"b4ce96e5c2241881","translatedAt":"2026-07-26T10:24:54.418Z"},"hi":{"title":"FAIRChem v2 UMA सामान्य मशीन लर्निंग एटॉमिक इंटरेक्शन पोटेंशियल जारी","summary":"FAIRChem v2 ने UMA सामान्य मशीन लर्निंग एटॉमिक इंटरएक्शन पोटेंशियल जारी किया, जो अणु रसायन, उत्प्रेरक और अकार्बनिक सामग्री के तीन मुख्य क्षेत्रों में एकल-बिंदु ऊर्जा पूर्वानुमान, ज्यामिति अनुकूलन, अणु गतिशीलता आदि कार्यों को निष्पादित कर सकता है। यह फ्रेमवर्क Hugging Face प्रमाणन के माध्यम से gated मॉडल वज़न प्राप्त करता है, और परमाणु संरचना और ट्रेजेक्टरी विश्लेषण को प्रबंधित करने के लिए Atomic Simulation Environment को एकीकृत करता है।","category":"技巧观点","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"FAIRChem v2 UMA सामान्य मशीन लर्निंग एटॉमिक इंटरेक्शन पोटेंशियल जारी - Aioga AI समाचार","description":"FAIRChem v2 ने UMA सामान्य मशीन लर्निंग एटॉमिक इंटरएक्शन पोटेंशियल जारी किया, जो अणु रसायन, उत्प्रेरक और अकार्बनिक सामग्री के तीन मुख्य क्षेत्रों में एकल-बिंदु ऊर्जा पूर्वानुमान, ज...","url":"https://www.aioga.com/hi/news/cms1ke8td00c6ro05i2sb5cn4/","contentTranslated":true,"sourceHash":"b4ce96e5c2241881","translatedAt":"2026-07-26T10:24:57.306Z"},"it":{"title":"Rilascio del potenziale interatomico di apprendimento automatico generale FAIRChem v2 UMA","summary":"FAIRChem v2 ha lanciato il potenziale atomico universale di machine learning UMA, in grado di eseguire previsioni di energia a singolo punto, ottimizzazione geometrica, dinamica molecolare e altri compiti attraverso i tre principali campi della chimica molecolare, catalisi e materiali inorganici. Questo framework ottiene pesi modello gated certificati tramite Hugging Face e integra Atomic Simulation Environment per gestire strutture atomiche e analisi delle traiettorie.","category":"技巧观点","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"Rilascio del potenziale interatomico di apprendimento automatico generale FAIRChem v2 UMA - Aioga Notizie IA","description":"FAIRChem v2 ha lanciato il potenziale atomico universale di machine learning UMA, in grado di eseguire previsioni di energia a singolo punto, ottimizzazione geometrica, dinamica mo...","url":"https://www.aioga.com/it/news/cms1ke8td00c6ro05i2sb5cn4/","contentTranslated":true,"sourceHash":"b4ce96e5c2241881","translatedAt":"2026-07-26T10:25:43.170Z"},"nl":{"title":"FAIRChem v2 UMA algemene machine learning interatomaire potentialen uitgebracht","summary":"FAIRChem v2 introduceert de UMA universele machine learning interatomische potentiëlen, die enkelpuntsenergievoorspelling, geometrie-optimalisatie, moleculaire dynamica en andere taken kunnen uitvoeren over drie grote gebieden: moleculaire chemie, katalyse en anorganische materialen. Dit framework verkrijgt gated modelgewichten via Hugging Face-certificering en integreert de Atomic Simulation Environment voor het beheren van atomaire structuren en trajectanalyse.","category":"技巧观点","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"FAIRChem v2 UMA algemene machine learning interatomaire potentialen uitgebracht - Aioga AI-nieuws","description":"FAIRChem v2 introduceert de UMA universele machine learning interatomische potentiëlen, die enkelpuntsenergievoorspelling, geometrie-optimalisatie, moleculaire dynamica en andere t...","url":"https://www.aioga.com/nl/news/cms1ke8td00c6ro05i2sb5cn4/","contentTranslated":true,"sourceHash":"b4ce96e5c2241881","translatedAt":"2026-07-26T10:25:37.436Z"},"tr":{"title":"FAIRChem v2 UMA Genel Amaçlı Makine Öğrenimi Atomlar Arası Potansiyel Yayınlandı","summary":"FAIRChem v2, UMA genel amaçlı makine öğrenimi atomlar arası potansiyelini sunar; bu, moleküler kimya, kataliz ve inorganik malzeme olmak üzere üç ana alanda tek nokta enerji tahmini, geometri optimizasyonu, moleküler dinamik gibi görevleri gerçekleştirebilir. Bu çerçeve, Hugging Face sertifikalı gated model ağırlıklarını kullanarak elde eder ve Atomik Simülasyon Ortamı ile atom yapıları ve iz analizi yönetimini entegre eder.","category":"技巧观点","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"FAIRChem v2 UMA Genel Amaçlı Makine Öğrenimi Atomlar Arası Potansiyel Yayınlandı - Aioga AI Haberleri","description":"FAIRChem v2, UMA genel amaçlı makine öğrenimi atomlar arası potansiyelini sunar; bu, moleküler kimya, kataliz ve inorganik malzeme olmak üzere üç ana alanda tek nokta enerji tahmin...","url":"https://www.aioga.com/tr/news/cms1ke8td00c6ro05i2sb5cn4/","contentTranslated":true,"sourceHash":"b4ce96e5c2241881","translatedAt":"2026-07-26T10:26:31.488Z"},"vi":{"title":"FAIRChem v2 UMA phát hành tiềm năng liên nguyên tử học máy chung","summary":"FAIRChem v2 ra mắt tiềm năng giữa các nguyên tử học máy UMA chung, có thể thực hiện dự đoán năng lượng một điểm, tối ưu hóa hình học và động lực học phân tử trên ba lĩnh vực chính: hóa học phân tử, xúc tác và vật liệu vô cơ. Khung này nhận được trọng số mô hình bị khóa thông qua chứng nhận của Hugging Face và tích hợp môi trường mô phỏng nguyên tử (Atomic Simulation Environment) để quản lý cấu trúc nguyên tử và phân tích quỹ đạo.","category":"技巧观点","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"FAIRChem v2 UMA phát hành tiềm năng liên nguyên tử học máy chung - Tin tức AI Aioga","description":"FAIRChem v2 ra mắt tiềm năng giữa các nguyên tử học máy UMA chung, có thể thực hiện dự đoán năng lượng một điểm, tối ưu hóa hình học và động lực học phân tử trên ba lĩnh vực chính:...","url":"https://www.aioga.com/vi/news/cms1ke8td00c6ro05i2sb5cn4/","contentTranslated":true,"sourceHash":"b4ce96e5c2241881","translatedAt":"2026-07-26T10:26:24.165Z"},"id":{"title":"FAIRChem v2 UMA Rilis Potensi Antaratom Pembelajaran Mesin Umum","summary":"FAIRChem v2 meluncurkan potensi antar-atom pembelajaran mesin universal UMA, yang dapat melakukan prediksi energi titik tunggal, optimisasi geometri, dinamika molekul, dan tugas lainnya di tiga bidang utama: kimia molekuler, katalisis, dan material anorganik. Kerangka ini memperoleh bobot model bersyarat melalui sertifikasi Hugging Face dan mengintegrasikan Atomic Simulation Environment untuk mengelola struktur atom dan analisis lintasan.","category":"技巧观点","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"FAIRChem v2 UMA Rilis Potensi Antaratom Pembelajaran Mesin Umum - Berita AI Aioga","description":"FAIRChem v2 meluncurkan potensi antar-atom pembelajaran mesin universal UMA, yang dapat melakukan prediksi energi titik tunggal, optimisasi geometri, dinamika molekul, dan tugas la...","url":"https://www.aioga.com/id/news/cms1ke8td00c6ro05i2sb5cn4/","contentTranslated":true,"sourceHash":"b4ce96e5c2241881","translatedAt":"2026-07-26T10:27:13.912Z"},"th":{"title":"FAIRChem v2 UMA เผยแพร่พลังงานระหว่างอะตอมโดยใช้การเรียนรู้ของเครื่องทั่วไป","summary":"FAIRChem v2 เปิดตัว UMA ศักยภาพระหว่างอะตอมสำหรับการเรียนรู้ของเครื่องทั่วไป สามารถทำการทำนายพลังงานจุดเดียว การปรับรูปทรงเรขาคณิต และพลศาสตร์โมเลกุลข้ามสามสาขาหลัก ได้แก่ เคมีโมเลกุล การเร่งปฏิกิริยา และวัสดุอนินทรีย์ เครือข่ายนี้ได้รับน้ำหนักของโมเดลแบบ gated ผ่านการรับรองจาก Hugging Face และรวม Atomic Simulation Environment สำหรับจัดการโครงสร้างอะตอมและการวิเคราะห์เส้นทาง","category":"技巧观点","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"FAIRChem v2 UMA เผยแพร่พลังงานระหว่างอะตอมโดยใช้การเรียนรู้ของเครื่องทั่วไป - ข่าว AI Aioga","description":"FAIRChem v2 เปิดตัว UMA ศักยภาพระหว่างอะตอมสำหรับการเรียนรู้ของเครื่องทั่วไป สามารถทำการทำนายพลังงานจุดเดียว การปรับรูปทรงเรขาคณิต และพลศาสตร์โมเลกุลข้ามสามสาขาหลัก ได้แก่ เคมีโมเล...","url":"https://www.aioga.com/th/news/cms1ke8td00c6ro05i2sb5cn4/","contentTranslated":true,"sourceHash":"b4ce96e5c2241881","translatedAt":"2026-07-26T10:27:18.458Z"},"pl":{"title":"FAIRChem v2 UMA wydanie uniwersalnego uczenia maszynowego potencjału międzyatomowego","summary":"FAIRChem v2 wprowadza uniwersalny międzyatomowy potencjał uczenia maszynowego UMA, który może wykonywać prognozowanie energii pojedynczego punktu, optymalizację geometrii, dynamikę molekularną i inne zadania w trzech głównych dziedzinach: chemii molekularnej, katalizie i materiałach nieorganicznych. Ta struktura uzyskuje wagę modelu typu gated za pośrednictwem certyfikacji Hugging Face i integruje Atomic Simulation Environment do zarządzania strukturami atomowymi i analizą trajektorii.","category":"技巧观点","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"FAIRChem v2 UMA wydanie uniwersalnego uczenia maszynowego potencjału międzyatomowego - Aioga Wiadomości AI","description":"FAIRChem v2 wprowadza uniwersalny międzyatomowy potencjał uczenia maszynowego UMA, który może wykonywać prognozowanie energii pojedynczego punktu, optymalizację geometrii, dynamikę...","url":"https://www.aioga.com/pl/news/cms1ke8td00c6ro05i2sb5cn4/","contentTranslated":true,"sourceHash":"b4ce96e5c2241881","translatedAt":"2026-07-26T10:28:02.534Z"}}}}