{"article":{"annotation":{"article_id":"arm-agi-cpu-agentic-benchmark-methodology-october-2026","available_at":"2026-10-13T10:04:14.741001Z","geography_revision":"ef6ee0010dda5fe7b240fd514749b43e214c57da73222d2e40ec83b276dd0977","id":"440a575b89d3aea407cd9d8167398ec24298dbebc8a6c738280c0468e2c03a7e","input_sha256":"de0365402e8b5b409e74d1372ca75428bfdf5faa143417984a4aae51b7355f1a","limitations":["Rule-derived subjects are candidates, not verified editorial labels or calibrated model probabilities.","Coverage is distinct from event location. Unknown country/city/coordinates remain absent; no automatic geocoding.","Article geometries keep their stated precision and methodology; reproducing them is not independent coordinate verification."],"method":"controlled-rules-v1","places":[],"rules_sha256":"fb5ad3c27c91cb840f7a0a05f7f64a5cb26ad0374409ff9457f277c7092e22b4","schema_version":1,"subjects":[{"concept_id":"medtop:13000000","confidence":null,"confidence_method":null,"evidence":[{"field":"category+topics","language":"en","note":"Preserved editorial category/topic; broad candidate only","source_urls":["https://newsroom.arm.com/news/arm-agi-cpu-agentic-benchmarks"]}],"method":"controlled-rules-v1","rule_id":"technology-broad","status":"candidate"}],"taxonomy_revision":"1f2db31d811fca07e99b16d402c10e2cc2965c9df1c7b41d9be348b40a24e7fb","taxonomy_version":"IPTC Media Topics 2026-Q2"},"api_url":"https://newsverum.com/api/v1/articles/arm-agi-cpu-agentic-benchmark-methodology-october-2026?lang=en","category":"technology","content":{"key_points":[],"paragraphs":["Arm introduced an agent-focused CPU evaluation method on October 13, with support from Fujitsu, Meta and OpenAI. The company said the same benchmark scripts ran on Arm AGI CPU and AMD EPYC servers, and that Signal65 independently verified the server-level measurements. The framework is intended to combine representative agent use cases, repeatable CPU measurements and checks against end-to-end outcomes.","The rack comparison Arm disclosed is a model derived from measured server performance, rather than a direct measurement of a complete rack. In that model, AGI CPU averaged 1.99 times the rack-level performance of EPYC 9755 across seven workloads; category averages were 1.7 for tooling, 2.0 for agent workflows and orchestration, and 2.2 for reasoning and planning. Arm separately reported up to 45% lower sandbox-creation latency against EPYC 9965, a different comparison system. These findings are limited to the workloads, server measurements and rack assumptions described by Arm."],"sections":[]},"corrections":[],"country":null,"data":{"blocks":[{"columns":[{"key":"measure","label":{"ar":"القياس","de":"Messgröße","en":"Measure","es":"Medición","fr":"Mesure","pt":"Medição","tr":"Ölçüm"},"language":"en","type":"text"},{"key":"value","label":{"ar":"القيمة","de":"Wert","en":"Value","es":"Valor","fr":"Valeur","pt":"Valor","tr":"Değer"},"type":"number","unit":"count"}],"description":{"ar":"عدد أحمال العمل التي قورنت في نمذجة مستوى الرف.","de":"Anzahl der im Rack-Modell verglichenen Workloads.","en":"Number of workloads compared in the rack-level modeling.","es":"Número de cargas comparadas en el modelado por rack.","fr":"Nombre de charges comparées dans la modélisation par rack.","pt":"Número de cargas comparadas na modelagem por rack.","tr":"Raf düzeyi modellemesinde karşılaştırılan iş yükü adedi."},"id":"workload-count","methodology":{"ar":"يذكر إعلان Arm سبعة أحمال عمل.","de":"Die Arm-Mitteilung nennt sieben Workloads.","en":"Arm’s announcement reports seven workloads.","es":"El anuncio de Arm informa de siete cargas.","fr":"L’annonce d’Arm mentionne sept charges.","pt":"O anúncio da Arm informa sete cargas.","tr":"Arm duyurusu yedi iş yükü bildiriyor."},"rowTranslations":[{"measure":{"ar":"أحمال العمل في نموذج الرف لدى Arm","de":"Workloads im Arm-Rackmodell","en":"Arm rack-model workloads","es":"Cargas del modelo por rack de Arm","fr":"Charges du modèle de rack d’Arm","pt":"Cargas do modelo por rack da Arm","tr":"Arm raf-modeli iş yükleri"}}],"rows":[{"measure":"Arm rack-model workloads","value":7}],"sourceUrls":["https://newsroom.arm.com/news/arm-agi-cpu-agentic-benchmarks"],"title":{"ar":"عدد أحمال العمل في معيار Arm","de":"Anzahl der Workloads im Arm-Benchmark","en":"Workload count in Arm’s benchmark","es":"Número de cargas del benchmark de Arm","fr":"Nombre de charges du benchmark Arm","pt":"Número de cargas no benchmark da Arm","tr":"Arm kıyaslamasındaki iş yükü sayısı"},"type":"table","updatedAt":"2026-10-13T09:31:52Z"},{"columns":[{"key":"measure","label":{"ar":"المقارنة","de":"Vergleich","en":"Comparison","es":"Comparación","fr":"Comparaison","pt":"Comparação","tr":"Karşılaştırma"},"language":"en","type":"text"},{"key":"value","label":{"ar":"الانخفاض","de":"Senkung","en":"Reduction","es":"Reducción","fr":"Réduction","pt":"Redução","tr":"Azalma"},"type":"number","unit":"percent"}],"description":{"ar":"الحد الأعلى لخفض زمن الاستجابة في إعلان Arm.","de":"Obergrenze der Latenzsenkung laut Arm-Mitteilung.","en":"Upper-bound latency reduction in Arm’s announcement.","es":"Reducción máxima de latencia del anuncio de Arm.","fr":"Réduction maximale de latence annoncée par Arm.","pt":"Redução máxima de latência informada pela Arm.","tr":"Arm açıklamasındaki üst sınır gecikme azalması."},"id":"sandbox-latency-reduction","methodology":{"ar":"45% حد أعلى مقابل EPYC 9965؛ أما مقارنة أداء الرف المنفصلة فتستخدم EPYC 9755.","de":"45 % ist eine Obergrenze gegenüber EPYC 9965; EPYC 9755 ist der separate Rack-Leistungsvergleich.","en":"45% is an upper bound against EPYC 9965; EPYC 9755 is the separate rack-performance comparator.","es":"45 % es un límite superior frente a EPYC 9965; EPYC 9755 es el comparador distinto del rendimiento por rack.","fr":"45 % est une borne supérieure face à EPYC 9965; EPYC 9755 est le comparateur distinct pour les performances par rack.","pt":"45% é um limite superior ante o EPYC 9965; o EPYC 9755 é o comparador distinto de desempenho por rack.","tr":"Yüzde 45 üst sınırdır ve EPYC 9965 karşılaştırmasına aittir; EPYC 9755 raf-performansı karşılaştırıcısıdır."},"rowTranslations":[{"measure":{"ar":"انخفاض زمن إنشاء بيئة التنفيذ المعزولة مقابل EPYC 9965 (حد أعلى)","de":"Senkung der Sandbox-Erstellungslatenz gegenüber EPYC 9965 (Obergrenze)","en":"Sandbox creation latency reduction versus EPYC 9965 (upper bound)","es":"Reducción de latencia al crear sandbox frente a EPYC 9965 (límite superior)","fr":"Réduction de latence de création du bac à sable face à EPYC 9965 (borne supérieure)","pt":"Redução da latência de criação do sandbox ante o EPYC 9965 (limite superior)","tr":"EPYC 9965 karşısında kum havuzu oluşturma gecikmesi azalması (üst sınır)"}}],"rows":[{"measure":"Sandbox creation latency reduction versus EPYC 9965 (upper bound)","value":45}],"sourceUrls":["https://newsroom.arm.com/news/arm-agi-cpu-agentic-benchmarks"],"title":{"ar":"نتيجة زمن بيئة التنفيذ المعزولة التي أعلنتها Arm","de":"Von Arm berichtetes Ergebnis zur Sandbox-Latenz","en":"Sandbox latency result reported by Arm","es":"Resultado de latencia de sandbox comunicado por Arm","fr":"Résultat de latence du bac à sable rapporté par Arm","pt":"Resultado de latência de sandbox informado pela Arm","tr":"Arm tarafından bildirilen kum havuzu gecikmesi sonucu"},"type":"table","updatedAt":"2026-10-13T09:31:52Z"}],"version":1},"data_downloads":[{"formats":{"csv":"https://newsverum.com/api/v1/articles/arm-agi-cpu-agentic-benchmark-methodology-october-2026/data/workload-count?lang=en&format=csv","json":"https://newsverum.com/api/v1/articles/arm-agi-cpu-agentic-benchmark-methodology-october-2026/data/workload-count?lang=en&format=json"},"id":"workload-count"},{"formats":{"csv":"https://newsverum.com/api/v1/articles/arm-agi-cpu-agentic-benchmark-methodology-october-2026/data/sandbox-latency-reduction?lang=en&format=csv","json":"https://newsverum.com/api/v1/articles/arm-agi-cpu-agentic-benchmark-methodology-october-2026/data/sandbox-latency-reduction?lang=en&format=json"},"id":"sandbox-latency-reduction"}],"data_types":["table"],"editorial_illustration":{"alt":"Conceptual illustration of technology, processors and connected information.","asset_id":"newsverum-v1-technology","avif_variants":[{"height":320,"url":"https://newsverum.com/assets/newsverum/illustrations/v1/technology-480.avif","width":480},{"height":480,"url":"https://newsverum.com/assets/newsverum/illustrations/v1/technology-720.avif","width":720},{"height":640,"url":"https://newsverum.com/assets/newsverum/illustrations/v1/technology-960.avif","width":960},{"height":800,"url":"https://newsverum.com/assets/newsverum/illustrations/v1/technology-1200.avif","width":1200},{"height":1024,"url":"https://newsverum.com/assets/newsverum/illustrations/v1/technology-1536.avif","width":1536}],"credit":"Newsverum","documentary":false,"height":1024,"kind":"symbolic-editorial-illustration","language":"en","scope":"topic","url":"https://newsverum.com/assets/newsverum/illustrations/v1/technology-1536.webp","variants":[{"height":320,"url":"https://newsverum.com/assets/newsverum/illustrations/v1/technology-480.webp","width":480},{"height":640,"url":"https://newsverum.com/assets/newsverum/illustrations/v1/technology-960.webp","width":960},{"height":1024,"url":"https://newsverum.com/assets/newsverum/illustrations/v1/technology-1536.webp","width":1536}],"width":1536},"event_id":"event-cd55d506-0a85-5a23-83c2-f3e761ea059e","event_period":"2026-10-13","event_period_basis":"event-or-reference-period","id":"arm-agi-cpu-agentic-benchmark-methodology-october-2026","image":null,"importance":0.6,"language":"en","modified_at":"2026-10-13T10:04:09+00:00","publication_basis":"publication-record","published_at":"2026-10-13T10:04:09+00:00","reading_minutes":1,"region":"world","section_illustration":{"alt":"Conceptual illustration of technology, processors and connected information.","asset_id":"newsverum-v1-technology","avif_variants":[{"height":320,"url":"https://newsverum.com/assets/newsverum/illustrations/v1/technology-480.avif","width":480},{"height":480,"url":"https://newsverum.com/assets/newsverum/illustrations/v1/technology-720.avif","width":720},{"height":640,"url":"https://newsverum.com/assets/newsverum/illustrations/v1/technology-960.avif","width":960},{"height":800,"url":"https://newsverum.com/assets/newsverum/illustrations/v1/technology-1200.avif","width":1200},{"height":1024,"url":"https://newsverum.com/assets/newsverum/illustrations/v1/technology-1536.avif","width":1536}],"credit":"Newsverum","documentary":false,"height":1024,"kind":"symbolic-editorial-illustration","language":"en","scope":"section","url":"https://newsverum.com/assets/newsverum/illustrations/v1/technology-1536.webp","variants":[{"height":320,"url":"https://newsverum.com/assets/newsverum/illustrations/v1/technology-480.webp","width":480},{"height":640,"url":"https://newsverum.com/assets/newsverum/illustrations/v1/technology-960.webp","width":960},{"height":1024,"url":"https://newsverum.com/assets/newsverum/illustrations/v1/technology-1536.webp","width":1536}],"width":1536},"slug":"arm-agi-cpu-agentic-benchmark-methodology-october-2026","sources":[{"checkedAt":"2026-10-13","name":"Arm Newsroom","publishedAt":"2026-10-13","url":"https://newsroom.arm.com/news/arm-agi-cpu-agentic-benchmarks"}],"summary":"Signal65 used server-level measurements to model rack-level results across seven workloads; Arm reported a 1.99x modeled average for AGI CPU.","text":"Signal65 used server-level measurements to model rack-level results across seven workloads; Arm reported a 1.99x modeled average for AGI CPU.\n\nArm introduced an agent-focused CPU evaluation method on October 13, with support from Fujitsu, Meta and OpenAI. The company said the same benchmark scripts ran on Arm AGI CPU and AMD EPYC servers, and that Signal65 independently verified the server-level measurements. The framework is intended to combine representative agent use cases, repeatable CPU measurements and checks against end-to-end outcomes.\n\nThe rack comparison Arm disclosed is a model derived from measured server performance, rather than a direct measurement of a complete rack. In that model, AGI CPU averaged 1.99 times the rack-level performance of EPYC 9755 across seven workloads; category averages were 1.7 for tooling, 2.0 for agent workflows and orchestration, and 2.2 for reasoning and planning. Arm separately reported up to 45% lower sandbox-creation latency against EPYC 9965, a different comparison system. These findings are limited to the workloads, server measurements and rack assumptions described by Arm.","title":"Arm sets out an agentic CPU method and AGI CPU benchmark results","topics":[],"translations":[{"api_url":"https://newsverum.com/api/v1/articles/arm-agi-cpu-agentic-benchmark-methodology-october-2026?lang=tr","language":"tr","url":"https://newsverum.com/tr/news/arm-agi-cpu-agentic-benchmark-methodology-october-2026"},{"api_url":"https://newsverum.com/api/v1/articles/arm-agi-cpu-agentic-benchmark-methodology-october-2026?lang=en","language":"en","url":"https://newsverum.com/en/news/arm-agi-cpu-agentic-benchmark-methodology-october-2026"},{"api_url":"https://newsverum.com/api/v1/articles/arm-agi-cpu-agentic-benchmark-methodology-october-2026?lang=ar","language":"ar","url":"https://newsverum.com/ar/news/arm-agi-cpu-agentic-benchmark-methodology-october-2026"},{"api_url":"https://newsverum.com/api/v1/articles/arm-agi-cpu-agentic-benchmark-methodology-october-2026?lang=fr","language":"fr","url":"https://newsverum.com/fr/news/arm-agi-cpu-agentic-benchmark-methodology-october-2026"},{"api_url":"https://newsverum.com/api/v1/articles/arm-agi-cpu-agentic-benchmark-methodology-october-2026?lang=es","language":"es","url":"https://newsverum.com/es/news/arm-agi-cpu-agentic-benchmark-methodology-october-2026"},{"api_url":"https://newsverum.com/api/v1/articles/arm-agi-cpu-agentic-benchmark-methodology-october-2026?lang=de","language":"de","url":"https://newsverum.com/de/news/arm-agi-cpu-agentic-benchmark-methodology-october-2026"},{"api_url":"https://newsverum.com/api/v1/articles/arm-agi-cpu-agentic-benchmark-methodology-october-2026?lang=pt","language":"pt","url":"https://newsverum.com/pt/news/arm-agi-cpu-agentic-benchmark-methodology-october-2026"}],"url":"https://newsverum.com/en/news/arm-agi-cpu-agentic-benchmark-methodology-october-2026"},"version":"1"}