According to NVIDIA's blog, Firebird — described in the post as "an emerging AI cloud" — has launched what the announcement calls the CIS region's largest AI factory in Armenia, establishing what NVIDIA terms "a new AI computing hub powered by NVIDIA accelerated computing and Dell Technologies high-performance AI infrastructure" (NVIDIA blog).
The opening ceremony drew government officials from multiple countries. NVIDIA's blog reports that Nikol Pashinyan, prime minister of the Republic of Armenia; Zhaslan Madiyev, deputy prime minister of the Republic of Kazakhstan; and David Allen, U.S. chargé d'affaires, a.i. in Armenia, all attended the event — a lineup that signals the project's cross-border political visibility beyond a purely commercial data-center opening (NVIDIA blog).
NVIDIA's blog states that Firebird plans to deploy more than 70,000 NVIDIA Rubin and Blackwell GPUs and 300 megawatts of AI infrastructure capacity in Armenia by the end of 2027, which the post frames as work meant to accelerate "the country's development as a center for AI research, advanced computing and innovation" (NVIDIA blog).
The facility is built on the NVIDIA DSX platform, which NVIDIA's blog describes as integrating "accelerated computing, networking, power and cooling as one codesigned system." NVIDIA states that with DSX, the site "can run up to 40% more GPUs on the same footprint," which the announcement says results in "more tokens per dollar" and more value extracted "from every megawatt of capacity" (NVIDIA blog). Separately, NVIDIA's blog confirms the factory runs on "NVIDIA's total AI factory platform — reference architecture, accelerated computing, networking and AI software — and deployed on Dell PowerEdge servers" (NVIDIA blog).
| Metric | Value | Attributed to |
|---|---|---|
| GPU deployment target (by end of 2027) | 70,000+ NVIDIA Rubin and Blackwell GPUs | NVIDIA blog |
| Power capacity target (by end of 2027) | 300 megawatts | NVIDIA blog |
| GPU density gain via DSX platform | Up to 40% more GPUs on the same footprint | NVIDIA blog |
| Regional infrastructure roadmap | Approximately 2 gigawatts (Armenia, Kazakhstan, additional markets) | NVIDIA blog |
| Delivery timeline | Just over six months | NVIDIA blog |
| Officials at opening ceremony | 3 (Armenia, Kazakhstan, U.S.) | NVIDIA blog |
NVIDIA's blog credits two named infrastructure partners for the Hrazdan site. Schneider Electric, the post states, "provides the power infrastructure supporting Firebird's AI factory in Hrazdan," supplying "medium- and low-voltage switchgear, three-phase uninterruptible power supply systems and rack enclosures" that NVIDIA says helped Firebird "meet its accelerated deployment schedule" (NVIDIA blog).
On the cooling side, NVIDIA's blog states that Vertiv "provided a cooling architecture combining chilled-water technology, advanced controls and Vertiv TrimCooler technology for efficient heat rejection," coordinated through Vertiv's iCOM CWM Chilled Water Manager, which the post describes as improving "visibility, efficiency and responsiveness as demand shifts with AI workloads" (NVIDIA blog).
NVIDIA's blog states the Armenia AI factory was "delivered in just over six months," a timeline the post says "demonstrates Firebird's ability to turn ambitious infrastructure plans into operational AI capacity with exceptional speed" (NVIDIA blog). That characterization of speed comes directly from NVIDIA's announcement rather than from an independent benchmark, and no comparison facility or industry-average build time is provided in the source material.
Beyond the Armenia site, NVIDIA's blog states that Firebird, "with NVIDIA's support," is pursuing "an approximately 2-gigawatt AI infrastructure roadmap spanning Armenia, Kazakhstan and additional markets." The same post reports that NVIDIA "intends to invest in the company, following an earlier investment by CoreWeave this year" (NVIDIA blog). No investment amount for either NVIDIA's intended stake or CoreWeave's earlier investment is disclosed in the source.
NVIDIA's blog identifies early customer demand coming from "AI-native companies including Perplexity, which is working with Firebird to access high-performance AI infrastructure for its AI agent platform and answer engine" (NVIDIA blog). No other named customers or contract terms are disclosed in the source material.
The facts NVIDIA's blog discloses trace a single narrative arc: a six-month build (NVIDIA blog) feeding into a much larger, multi-year target — 70,000-plus GPUs and 300 megawatts by the end of 2027 (NVIDIA blog) — that itself sits inside a still-larger, longer-horizon regional plan of roughly 2 gigawatts across Armenia, Kazakhstan and other markets (NVIDIA blog). The DSX platform's claimed 40% density gain (NVIDIA blog) is presented as the mechanism that lets Firebird pursue this scale-up without a proportional increase in footprint, while the named suppliers — Schneider Electric on power and Vertiv on cooling (NVIDIA blog) — are credited specifically with enabling the compressed six-month schedule. Firebird's investor lineup, moving from CoreWeave's earlier investment to NVIDIA's stated intent to invest (NVIDIA blog), lines up with the presence of government officials from three jurisdictions at the opening (NVIDIA blog), and with an early customer, Perplexity, described as seeking capacity for AI agent and answer-engine workloads (NVIDIA blog). All of these figures and quotes originate from a single NVIDIA blog post, so the article reports what NVIDIA and Firebird say about the facility rather than independently verified performance or financial data.
According to NVIDIA's blog, Firebird plans to deploy more than 70,000 NVIDIA Rubin and Blackwell GPUs and 300 megawatts of AI infrastructure capacity in Armenia by the end of 2027.
NVIDIA's blog states the facility was delivered in just over six months.
NVIDIA's blog reports that NVIDIA intends to invest in Firebird, following an earlier investment by CoreWeave this year; no investment amounts are disclosed.
According to NVIDIA's blog, Schneider Electric supplied power infrastructure including medium- and low-voltage switchgear, UPS systems and rack enclosures, while Vertiv supplied the chilled-water cooling architecture coordinated by its iCOM CWM Chilled Water Manager.
According to TechNews and CNA, Winbond Electronics (華邦電) GM Chen Pei-ming warned on August 6, 2026 that 2027 memory shortages will exceed 2026's. CMS revenue rose 78% and Flash revenue rose 65% quarter-over-quarter, backed by NT$39.5 billion in 2026 capex, up over 70-fold year-on-year.
According to Technews, MoneyDJ, and CTEE reports dated August 6, 2026, TSMC (台積電) is holding roughly $1 billion (about NT$32 billion) worth of Apple (蘋果) A20 Pro processors that cannot complete back-end packaging because of a DRAM shortage. TSMC's CFO had already flagged rising inventory days on July 16, and Apple sources memory mainly from Micron, plus SK Hynix and Samsung.
According to Technews, also reported by CNA and ETtoday, the US FCC is drafting a measure to ban imports of new Chinese-made optical transceivers, with officials hoping to finalize it this year, though it could still be revised or shelved. The news drove same-day share drops of 7.27% for Innolight (中際旭創) and 5.29% for Eoptolink (新易盛), while Beijing's foreign ministry called the move protectionist.