Semiconductor News · کورتەی دابینکردن

ASUS Adds 64 GB Ascent GX10 to the DGX Spark Lineup: What It Means for Local AI Workstation Sourcing

ASUS launches a 64 GB variant of the Ascent GX10 compact AI supercomputer built on NVIDIA DGX Spark, targeting more developers running 30-35B class models.

بڵاوکراوەتەوە نوێکراوەتەوە 4 خولەک خوێندنەوە ✓ ئامادەی AI ھاوبەشکردن

بە کورتی — سێ ڕستە

  • ASUS announced a new 64 GB configuration of the Ascent GX10 alongside the existing 128 GB model (S1).
  • The new SKU is built on the NVIDIA DGX Spark platform and runs the full NVIDIA AI software stack (S1).
  • The 64 GB unit is sized for 30-35B class open-source models, while two units cluster to 128 GB pooled memory at around 1.7X throughput (S1).

وەرگیراوە لە key_takeaways (خانەی ڕاستەقینە، زۆرترین 3 تۆمار). ڕاستی و شیکاری بە جیا نیشانە کراون.

ASUS Adds 64 GB Ascent GX10 to the DGX Spark Lineup: What It Means for Local AI Workstation Sourcing — Semiconductor News news briefEditorial cover for a Semiconductor News brief. Chip motif; headline "ASUS Adds 64 GB Ascent GX10 to the DGX Spark Lineup: What It Means for Local A…". Reported key numbers: 64 GB; 128 GB; 128 GB. 1 cited source. A one-time entrance animation stops under reduced motion.SEMICONDUCTOR NEWS · NEWS BRIEFASUS Adds 64 GB Ascent GX10 to the DGX SparkLineup: What It Means for Local A…REPORTED KEY NUMBERS64 GB128 GB128 GB1 sourceXAIPART
Cover image generated from data already shown on this page: the category motif, the headline, and reported key numbers. No decorative illustration; no third-party image.

لە وەڵامێکی AIەوە هاتوویت — ئەمە ئەسڵی پشتڕاستکراوەیە

لەوانەیە پوختەیەکی AI بنەمای داواکارییەکانی لەدەستدابێت. پێشبینییە ڕەسەنەکە، بەشەکانی تووشبوو و هەموو ئاماژەکان لەم پەڕەیەدان، هەریەکەیان بەرەو سەرچاوەکەی دەگەڕێتەوە.

پێشبینی XAIPART

ئەم کورتەیە پێشبینییەکی هەڵوەشێنراوەی تێدایە.

بەشەکانی تووشبوو

ھیچ گونجانێک لە کاتالۆگدا بۆ ئەم چیرۆکە نییە — ھیچ پیشان نادرێت لە جیاتی دەرھێنانی گونجان.

Reported by TechPowerUp on 2026-10-05. Summarized and analyzed by XAIPART Editorial.

What happened

ASUS announced a new 64 GB configuration of the ASUS Ascent GX10, expanding the lineup alongside the existing 128 GB model.S1 The compact AI supercomputer is aimed at developers, researchers and data scientists working on local agentic AI development.S1

According to the source, the new SKU is built on the NVIDIA DGX Spark platform and runs the full NVIDIA AI software stack.S1 ASUS positions the 64 GB unit as sharing the same petaflop-scale architecture as the 128 GB Ascent GX10, but with a memory configuration targeted at what the company calls today's efficient open-source models.S1

The release notes that the 64 GB system is sized to "build and run today's most capable 30-35B class models and autonomous agents," and that two units can be clustered using the NVIDIA Sync Cluster Assistant to reach a pooled 128 GB of memory with "around 1.7X the throughput."S1 The brief is framed inside ASUS's broader "Ubiquitous AI. Incredible Possibilities." strategy and is positioned for "the much larger early-majority audience of developers, researchers and AI-curious teams."S1

Key numbers
MetricReported valueSource
New SKU memory64 GB[S1]
Existing SKU memory128 GB[S1]
Target model class30-35B class open-source models[S1]
Two-unit cluster memory (pooled)128 GB[S1]
Two-unit cluster throughput gain~1.7X (around)[S1]
List price (Ascent GX10 64 GB)Not disclosed—
Ship date / availabilityNot disclosed—

Background timeline

  1. - ASUS announces a new 64 GB configuration of the Ascent GX10 alongside the existing 128 GB model.S1
  2. not disclosed - Prior launch date of the original 128 GB ASUS Ascent GX10 was not disclosed by the source.

XAIPART take

Technical impact

The 64 GB SKU keeps the DGX Spark silicon and full NVIDIA AI software stack intact, so toolchain compatibility with the 128 GB version should carry over unchanged for ISVs and in-house MLOps teams. DXAIPART expects the practical ceiling to land at 30-35B parameter models in FP4/FP8 quantizations rather than full-precision BF16 runs, given 64 GB of unified memory on this class of platform.

DXAIPART analysis - The two-unit cluster path to a pooled 128 GB via the NVIDIA Sync Cluster Assistant is the more interesting scaling story: teams that outgrow one box can add a second without a forklift upgrade, which is closer to a node-add than a typical GPU server refresh.

Market impact

DXAIPART analysis - Adding a 64 GB SKU below the 128 GB Ascent GX10 broadens the addressable buyer base from frontier-scale local AI teams into the early-majority developer segment the ASUS press text explicitly names. For procurement, this creates a clearer two-step roadmap: start at 64 GB, cluster to 128 GB when workloads demand it, instead of committing to a higher-tier box on day one.

DXAIPART analysis - Distributors and resellers should expect the 64 GB unit to become the higher-velocity SKU and the 128 GB to remain the halo configuration, mirroring typical workstation tiering.

Action

DXAIPART analysis - Confirm pricing, channel stocking and ASUS partner allocation for the Ascent GX10 64 GB before quoting on customer BOMs; the source did not disclose MSRP or ship dates.

DXAIPART analysis - For teams currently shortlisting DGX Spark-class workstations, validate that their target 30-35B open-source models fit within 64 GB of unified memory at the planned precision, and size the second-unit cluster option into the rack, power and networking plan rather than treating it as a future afterthought.

XAIPART Forecast

XAIPART analysisDXAIPART expects that within 9 months of the 2026-10-05 announcement, ASUS or its channel partners will publish a US-dollar price tag for the Ascent GX10 64 GB SKU; if a published single-unit price does not appear by 2027-07-05, this forecast is wrong.

Facts are as reported by the cited source. The design-in, sourcing and forward-looking statements above are XAIPART's analysis, not the source's claims. Uncertainty: ASUS's launch announcement did not disclose pricing, ship dates, regional channel allocation, or a confirmable first-availability date; the 1.7X cluster throughput figure is described by the source only as "around" 1.7X and is single-source. XAIPART analysis

Corrections: none at publication. This brief will be updated with a timestamp if the source corrects its report.

جێگرەوەکان و چرخەی ژیان

جێگرەوە تۆمارکراوەکان — مەودای کاتالۆگ، نەک ڕیزبەندی بازاڕ
بەشپەیوەندیخولگەی ژیانبنەما
— ھیچ بەدیلێک لە تۆماردا نییە پێویستی بە داتا ھەیە — پێویستی پێیەتی match_type پێش ئەوەی هیچ نیشانەیەکی پێشوو دەربکەوێت

نیشانەکانی «پێشوو» و «بەدیل» تەنھا لە جۆرێکی گونجانی تۆمارکراو، دۆخی چرخەی ژیان، یان دەقی سەرچاوەوە دێن. لەو شوێنەی ھیچ گونجانێک تۆمار نەکراوە، ھیچ نیشانەیەک پیشان نادرێت. لیستەکانی خۆمان ئاماژەیەکی بازاڕ نین.

چی چاودێری بکەیت

پێشبینی ڕەتکراوە — سنوور، ماوە و دۆخی ئێستا. پێشبینییە ڕەتکراوەکان بە دیار دەمێننەوە.
داواکاریسنوورلەلایەندۆخ
DXAIPART expects that within 9 months of the 2026-10-05 announcement, ASUS or its channel partners will publish a US-dollar price tag for the Ascent GX10 64 GB SKU a published single-unit price does not appear by 2027-07-05 Pending

ئەو پێشبینییانەی هەڵە دەردەچن بە دیار دەمێننەوە و نیشان دەدرێن. ئێمە لەناویان نابەین.

دیمەنی کێبڕکێ

  • 27953 لیستەی زیندوو لەم پۆلەدا لەسەر XAIPART
  • 113 بەرهەمهێنەرانی نوێنەرایەتیکراو
  • 52 بەرهەمهێنەرانی ئاماژەپێکراو لە ئەرشیفی هەواڵەکانمان لە ٣٠ ڕۆژی ڕابردوودا

ژمارەکان تەنها کاتالۆگی XAIPART نیشان دەدەن — پشکی بازاڕ نین.

بەشەکانی تووشبوو

ھیچ گونجانێک لە کاتالۆگدا نییە. ئەم هەواڵە بۆ هیچ پارچەیەک کە ئێستا لە XAIPART لیستکراوە ناگونجێت. ئێمە هاوتایی نادۆزینەوە، و لیستێکی شوێنگرەوە نیشان نادەین.

ئەمە لە کوێی چیرۆکەکەدایە

Event chain · as of 2026-11-19T01:30:53+08:00 · nodes link to published articles on this site.
Event chain — asus+semiconductor-newsChain nodes: 2026-10-11 ASUS launches a 64 GB variant of the Ascent GX10 compact AI supercomputer built on NVIDIA DGX Spark, targeting more developers running 30-35B class models.; 2026-11-03 ASUS's updated specs for the PCIe-slot-powered XG Core Radeon RX 9060 XT LP cap the whole unit at 75 W, forcing clocks far below the desktop card — relevant to eGPU BOM and thermal budgets.; 2026-11-18 ASUS presented a multi-platform AI infrastructure blueprint at SC26, including AMD EPYC 9006 systems, AMD Instinct accelerators and rack-scale systems already entering volume shipment.; 2026-11-19 ASUS has released the ProArt GR1X mini PC with 64 GB and 128 GB unified-memory options, giving buyers a Windows platform for local AI agents and creative workloads.. As of 2026-11-19T01:30:53+08:00. Open forecast: ASUS launches a 64 GB variant of the Ascent GX10 compact AI supercompute — pending Open forecast: ASUS's updated specs for the PCIe-slot-powered XG Core Radeon RX 9060 XT — pending.2026-10-11ASUS launches a 64our brief ↗this brief2026-11-03ASUS's updated speour brief ↗2026-11-18ASUS presented a mour brief ↗2026-11-19ASUS has released our brief ↗Open forecast: ASUS launches a 64 GB variant of the Ascent GX10 compact AI supercompute — pendingOpen forecast: ASUS's updated specs for the PCIe-slot-powered XG Core Radeon RX 9060 XT — pending
  1. asus-adds-64-gb-ascent-gx10-to-the-dgx-spark-lineup-what-it-means-for-local-ai-workstation-sourcing2026-10-11 · product launch
  2. asus-xg-core-radeon-rx-9060-xt-lp-trades-clocks-for-a-75-w-slot-powered-dock-design2026-11-03
  3. asus-sc26-ai-infrastructure-line-targets-amd-intel-and-nvidia-deployments2026-11-18
  4. proart-gr1x-availability-puts-128-gb-rtx-spark-mini-pcs-on-the-local-ai-buying-list2026-11-19 · product launch
  1. ASUS launches a 64 GB variant of the Ascent GX10 compact AI supercomputer built on NVIDIA DGX Spark, targeting more developers running 30-35B class models. product launch trade
  2. ASUS's updated specs for the PCIe-slot-powered XG Core Radeon RX 9060 XT LP cap the whole unit at 75 W, forcing clocks far below the desktop card — relevant to eGPU BOM and thermal budgets. trade
  3. ASUS presented a multi-platform AI infrastructure blueprint at SC26, including AMD EPYC 9006 systems, AMD Instinct accelerators and rack-scale systems already entering volume shipment. trade
  4. ASUS has released the ProArt GR1X mini PC with 64 GB and 128 GB unified-memory options, giving buyers a Windows platform for local AI agents and creative workloads. product launch trade

زنجیرەکە تەنها لە کورتە بڵاوکراوەکان دروست دەکرێت — گرێیەک کە کورتەی نییە کێشراوە نییە.

فشاری تێچوو بەپێی ڕەهەند

Source: FRED · Generated · reference output z and composite are algorithm outputs, not a fetch snapshot

Cost pressure by dimension — composite stateDiverging bars for three dimensions around a zero midpoint. Input Cost has no reading; Supply Demand z 0.0276, neutral; Fx Energy z -0.0087, neutral. Composite index 0.1, band neutral.0Input costno readingSupply/demandz +0.0276 · neutral48 dFX & energyz −0.0087 · neutral5 dcomposite +0.1 · neutral−100 · −40 · −15 · 0 · +15 · +40 · +100
Underlying values — copyable for verification.
DimensionFRED series IDUnitLagAs of
Input costPCOPPUSDM, PALUMUSDM, PNICKUSDMUSD/mt52 d
Supply/demandPCU334413334413, IPG3344Sindex(1982=100) / index(2017=100)48 d
FX & energyDEXCHUS, DCOILBRENTEU, DHHNGSPrate / USD/bbl / USD/MMBtu5 d

No free authoritative interface for: gold · silver · rare earth · polysilicon · lithium carbonate · market share — listed rather than approximated.

ئەمە چۆن بەرهەم هێنرا

بەڵگەی پرۆسە — ھێنان، ڕاستی سەرچاوە، مۆدێل و دەروازە
شوێنپێی پشکنین بۆ ئەم کورتەنامەیە
Source fetched TechPowerUp · www.techpowerup.com · 2,033 chars · fetch status success
Source sanity (S1) ok — passed before any model call (no s1_ block recorded)
Model / prompt MiniMax-M3 · prompt Blog Agent - News Summary (v2.8-final) · 6,283 tokens · 23.4s · run completed
Gate verdict hold · 6/6 applicable checks passed · spec v2.9 · C10 ratio 0.064879 · C10 rounds 0
Time contract 1 <time datetime> values in this brief — 0 violations (+08:00 or date-only)

هەر ڕیزێکی سەرەوە لەلایەن پرۆسێسەکەوە بەرهەم دەهێنرێت، نەک بە دەست نووسرابێت.

نمرەی سەروەرێتی

ئەنجامی دەروازەی دیاریکراو — PHP ی خاوێن، هیچ مۆدێلێک لە بڕیاردانەکەدا بەشدار نییە
شیکردنەوەی نمرەی دەروازە 6 پشکنینەکانی جێبەجێکراو، 6دەرپەڕیوە: 100 لەسەد. Hold items: C13 (falsifiable forecast) — sent to human review. Gate spec: v2.9 (S0 + S1 + C1–C20) · Source affinity (C20): 0.6742. پشکنینەکانی جێبەجێکراو تێپەڕێنراون 6 / 6 = 100% · بڕیار: hold Hold items: C13 (falsifiable forecast) — sent to human review Gate spec: v2.9 (S0 + S1 + C1–C20) · Source affinity (C20): 0.6742 دەروازەکە بڕیار لە بڵاوکردنەوە دەدات. نمرەی وردبینی سەربەخۆ تەنها ڕاوێژییە.
هەڵەیەکت دۆزیوەتەوە؟ پێمان بڵێ چی هەڵەیە — ڕاستکردنەوەکان لەگەڵ دەقی ڕەسەن بڵاودەکرێنەوە و تۆمار دەکرێن. بە نهێنی وتارەکان دەستکاری ناکەین.

بۆچوونەکان

یەکەم کەس بە بۆچوون بنووسە.

سەرچاوەکان

  1. ASUS Announces New 64GB Configuration of the Ascent GX10 — https://www.techpowerup.com/353392/asus-announces-new-64gb-configuration-of-the-ascent-gx10 · rss · سەردانکراوە

ڕاستییەکان بە S1/S2 نیشان دەدرێن و بۆ سەرچاوەی بەستەردراو دەگەڕێنەوە. ئەو دەقانەی نیشانکراون شیکاری بیرکردنەوەی خۆی XAIPARTن، جیاکراوە لە ڕاستییەکانی ڕاپۆرتکراو. بەستەرەکانی سەرچاوە شوێن دەکرێن نەک nofollow — دانانی سەرچاوە بە مەبەستە.