{"version":"1","article":{"id":"nvidia-dgx-spark-64gb-local-ai-october-2026","slug":"nvidia-dgx-spark-64gb-local-ai-october-2026","language":"en","url":"https://haberpaper.com/en/news/nvidia-dgx-spark-64gb-local-ai-october-2026","api_url":"https://haberpaper.com/api/v1/articles/nvidia-dgx-spark-64gb-local-ai-october-2026?lang=en","title":"NVIDIA sets Oct. 23 start for 64GB DGX Spark configuration","summary":"NVIDIA said Acer, ASUS, Dell, Gigabyte, HP and MSI will offer a 64GB unified-memory DGX Spark configuration from Oct. 23, starting at $4,999. In the company’s test, two systems reached up to 1.7 times the performance of one system on Qwen 3.8 27B.","category":"business","region":"world","topics":[],"published_at":"2026-10-08T01:26:48+03:00","modified_at":"2026-10-08T01:26:48+03:00","publication_basis":"publication-record","event_period":"2026-10-02","event_period_basis":"event-or-reference-period","sources":[{"name":"NVIDIA Blog","url":"https://blogs.nvidia.com/blog/local-ai-dgx-spark-64gb-sync/","publishedAt":"2026-10-02","checkedAt":"2026-10-08"}],"translations":[{"language":"tr","url":"https://haberpaper.com/tr/news/nvidia-dgx-spark-64gb-local-ai-october-2026","api_url":"https://haberpaper.com/api/v1/articles/nvidia-dgx-spark-64gb-local-ai-october-2026?lang=tr"},{"language":"en","url":"https://haberpaper.com/en/news/nvidia-dgx-spark-64gb-local-ai-october-2026","api_url":"https://haberpaper.com/api/v1/articles/nvidia-dgx-spark-64gb-local-ai-october-2026?lang=en"},{"language":"ar","url":"https://haberpaper.com/ar/news/nvidia-dgx-spark-64gb-local-ai-october-2026","api_url":"https://haberpaper.com/api/v1/articles/nvidia-dgx-spark-64gb-local-ai-october-2026?lang=ar"},{"language":"fr","url":"https://haberpaper.com/fr/news/nvidia-dgx-spark-64gb-local-ai-october-2026","api_url":"https://haberpaper.com/api/v1/articles/nvidia-dgx-spark-64gb-local-ai-october-2026?lang=fr"},{"language":"es","url":"https://haberpaper.com/es/news/nvidia-dgx-spark-64gb-local-ai-october-2026","api_url":"https://haberpaper.com/api/v1/articles/nvidia-dgx-spark-64gb-local-ai-october-2026?lang=es"}],"data_types":["events"],"data":{"version":1,"blocks":[{"id":"product-timeline","type":"events","title":{"tr":"Yapılandırma, fiyat ve üretici testinin kapsamı","en":"Configuration, price and scope of vendor testing","ar":"التكوين والسعر ونطاق اختبار الشركة","fr":"Configuration, prix et portée des tests du fabricant","es":"Configuración, precio y alcance de las pruebas del fabricante"},"description":{"tr":"NVIDIA’nın şirket duyurusundaki teknik özellikler ve bulunabilirlik takvimi; gerçekleşmiş satış veya bağımsız performans testi değildir.","en":"Specifications and availability schedule in NVIDIA’s company announcement; not completed sales or independent performance testing.","ar":"المواصفات وجدول التوفر في إعلان NVIDIA؛ وليست مبيعات مكتملة أو اختبار أداء مستقلا.","fr":"Spécifications et calendrier de disponibilité annoncés par NVIDIA; il ne s’agit ni de ventes réalisées ni d’un test indépendant.","es":"Especificaciones y calendario de disponibilidad anunciados por NVIDIA; no son ventas realizadas ni una prueba independiente."},"methodology":{"tr":"Üretici duyurusundaki özellik, liste fiyatı ve takvim aşama etiketleriyle derlenmiştir. Performans NVIDIA’ya atfedilir; duyuru, planlanan satış ve test gerçekleşmiş teslimattan ayrıdır.","en":"Vendor-announced specifications, list price and schedule are compiled with stage labels. Performance is attributed to NVIDIA; announcement, planned sale and test are distinct from completed delivery.","ar":"تجميع للمواصفات والسعر والجدول كما وردت في إعلان الشركة مع تحديد المرحلة. وتنسب ادعاءات الأداء إلى NVIDIA مع الفصل بين الإعلان والبيع المخطط والاختبار والتسليم المكتمل.","fr":"Compilation des spécifications, du prix et du calendrier annoncés avec distinction des étapes. Les performances sont attribuées à NVIDIA; annonce, vente prévue et test sont séparés de la livraison réalisée.","es":"Compilación de especificaciones, precio y calendario anunciados con etapas diferenciadas. El rendimiento se atribuye a NVIDIA; anuncio, venta prevista y prueba se distinguen de la entrega efectiva."},"sourceUrls":["https://blogs.nvidia.com/blog/local-ai-dgx-spark-64gb-sync/"],"updatedAt":"2026-10-08T00:55:15+03:00","items":[{"date":"2026-10-02","label":{"tr":"NVIDIA 64 GB yapılandırmasını duyurdu","en":"NVIDIA announced the 64GB configuration","ar":"أعلنت NVIDIA تكوين 64 غيغابايت","fr":"NVIDIA a annoncé la configuration 64 Go","es":"NVIDIA anunció la configuración de 64 GB"},"description":{"tr":"Yeni SKU ve üretici kanalları açıklandı.","en":"The new SKU and manufacturer sales channels were announced.","ar":"أُعلن عن النسخة الجديدة وقنوات البيع لدى الشركات المصنعة.","fr":"Le nouveau SKU et les canaux de vente des fabricants ont été annoncés.","es":"Se anunciaron el nuevo SKU y los canales de venta de los fabricantes."},"status":"occurred"},{"date":"2026-10-23","label":{"tr":"Planlanan satış başlangıcı","en":"Planned sales start","ar":"بدء البيع المخطط","fr":"Début des ventes prévu","es":"Inicio de venta previsto"},"description":{"tr":"Duyuruya göre ortak kanallarda 4.999 ABD doları başlangıç fiyatı.","en":"Announcement lists a USD 4,999 starting price through partners.","ar":"يذكر الإعلان سعرا ابتدائيا قدره 4,999 دولارا عبر الشركاء.","fr":"L’annonce indique un prix de départ de 4 999 USD chez les partenaires.","es":"El anuncio indica un precio inicial de 4.999 USD en los socios."},"status":"scheduled"}]}]},"data_compilations":[{"id":"product-timeline","creator":{"@type":"Organization","@id":"https://haberpaper.com/#organization","name":"Haber Paper","url":"https://haberpaper.com/"},"license":{"@type":"CreativeWork","@id":"https://haberpaper.com/en/data-terms/v1#license","name":"Haber Paper Original Data Compilation License 1.0","url":"https://haberpaper.com/en/data-terms/v1"},"rights_scope":"Haber Paper original data arrangement, labels and methodology notes only; underlying source rights and permissions remain separate.","source_license_url":null}],"data_downloads":[{"id":"product-timeline","formats":{"json":"https://haberpaper.com/api/v1/articles/nvidia-dgx-spark-64gb-local-ai-october-2026/data/product-timeline?lang=en&format=json","csv":"https://haberpaper.com/api/v1/articles/nvidia-dgx-spark-64gb-local-ai-october-2026/data/product-timeline?lang=en&format=csv"}}],"content":{"paragraphs":["NVIDIA announced on Oct. 2 a new DGX Spark configuration with 64GB of unified memory. It retains the company’s Grace Blackwell-based GB10 superchip, DGX OS and NVIDIA software stack, while offering less memory than the existing 128GB model. The manufacturer says the product will go on sale Friday, Oct. 23 through Acer, ASUS, Dell, Gigabyte, HP and MSI, at a starting price of $4,999. This is the announced starting retail price, not a business procurement rate or total cost of ownership.","NVIDIA says one 64GB system can run models with up to 100 billion parameters on device. Connecting two systems directly brings combined unified memory to 128GB; the company says the configuration can support models up to 200 billion parameters. In NVIDIA’s Qwen 3.8 27B test, two clustered 64GB systems delivered up to 1.7 times the performance of one system. That figure is the company’s result for a particular model and setup; it does not establish the same acceleration across models or an independent comparison.","The configuration is aimed at developers, researchers and users experimenting with local AI. NVIDIA highlights inference, fine-tuning, data science and agent workflows on the device without renting a cloud instance for every task. Clustering two units may simplify network and software configuration, but doubling memory alone does not determine application latency, model quality or cost. Results depend on the model, context length, data transfers and workload concurrency.","Acer, ASUS, Dell, Gigabyte, HP and MSI were named as sales partners; NVIDIA says the 64GB SKU will be offered through those manufacturers and will not replace the 128GB version. This channel distinction separates product availability from NVIDIA’s own hardware sales strategy. The $4,999 starting price applies to the configuration scheduled for Oct. 23; the announcement does not specify taxes, regional prices or inventory.","For organizations assessing AI investment, the economics of local inference depend on utilization. A device may keep sensitive data local and reduce variable cloud consumption, while purchase, electricity, cooling, operations and updates remain user costs. The announcement gives no sales forecast, customer count or total savings calculation versus cloud services. It should therefore be read as a priced product configuration and a vendor performance claim, not realized market demand or verified return on investment."],"key_points":[],"sections":[]},"text":"NVIDIA said Acer, ASUS, Dell, Gigabyte, HP and MSI will offer a 64GB unified-memory DGX Spark configuration from Oct. 23, starting at $4,999. In the company’s test, two systems reached up to 1.7 times the performance of one system on Qwen 3.8 27B.\n\nNVIDIA announced on Oct. 2 a new DGX Spark configuration with 64GB of unified memory. It retains the company’s Grace Blackwell-based GB10 superchip, DGX OS and NVIDIA software stack, while offering less memory than the existing 128GB model. The manufacturer says the product will go on sale Friday, Oct. 23 through Acer, ASUS, Dell, Gigabyte, HP and MSI, at a starting price of $4,999. This is the announced starting retail price, not a business procurement rate or total cost of ownership.\n\nNVIDIA says one 64GB system can run models with up to 100 billion parameters on device. Connecting two systems directly brings combined unified memory to 128GB; the company says the configuration can support models up to 200 billion parameters. In NVIDIA’s Qwen 3.8 27B test, two clustered 64GB systems delivered up to 1.7 times the performance of one system. That figure is the company’s result for a particular model and setup; it does not establish the same acceleration across models or an independent comparison.\n\nThe configuration is aimed at developers, researchers and users experimenting with local AI. NVIDIA highlights inference, fine-tuning, data science and agent workflows on the device without renting a cloud instance for every task. Clustering two units may simplify network and software configuration, but doubling memory alone does not determine application latency, model quality or cost. Results depend on the model, context length, data transfers and workload concurrency.\n\nAcer, ASUS, Dell, Gigabyte, HP and MSI were named as sales partners; NVIDIA says the 64GB SKU will be offered through those manufacturers and will not replace the 128GB version. This channel distinction separates product availability from NVIDIA’s own hardware sales strategy. The $4,999 starting price applies to the configuration scheduled for Oct. 23; the announcement does not specify taxes, regional prices or inventory.\n\nFor organizations assessing AI investment, the economics of local inference depend on utilization. A device may keep sensitive data local and reduce variable cloud consumption, while purchase, electricity, cooling, operations and updates remain user costs. The announcement gives no sales forecast, customer count or total savings calculation versus cloud services. It should therefore be read as a priced product configuration and a vendor performance claim, not realized market demand or verified return on investment.","corrections":[],"editorial_illustration":{"kind":"symbolic-editorial-illustration","documentary":false,"scope":"section","asset_id":"business","url":"https://haberpaper.com/assets/illustrations/business-1536.webp","width":1536,"height":1024,"alt":"Symbolic industrial workshop with machinery, a warehouse and a crane.","language":"en"},"section_illustration":{"kind":"symbolic-editorial-illustration","documentary":false,"scope":"section","asset_id":"business","url":"https://haberpaper.com/assets/illustrations/business-1536.webp","width":1536,"height":1024,"alt":"Symbolic industrial workshop with machinery, a warehouse and a crane.","language":"en"}}}