ALERT🚨 GB300 NVL72 NVLINK BACKPLANE ENABLES 13x BETTER PERF PER DOLLAR THAN HOPPER ON AGENTIC INFERENCE Through extreme co-design, GB300 NVL72 uses an innovative copper backplane to expand the scale-up domain to 72 versus 8 on Hopper. This allows optimizations like @vllm_project wide expert parallelism, which allows each GPU to have only a subset of experts.

NVIDIA GB300 NVL72 IS 7X BETTER 💰💰 PERF PER DOLLAR 💰💰 THAN H200 on long context agentic workloads due to disagg prefill and wide expert parallelism optimizations which take advantage of NVL72 copper backplane.

Fentanyl Grade Compute: Why a GB300 Rack Will Out-Price Blow by Weight in 2030 A GB300 NVL72 weighs roughly 1,580 kg fully populated and recent purchase orders put it at $5M per rack. That is $3,165/kg. Strip out the 1.5 tons of busbar, manifold and coolant and the GPU packages alone are well into gold territory, but we're pricing the rack, because that's what you actually take delivery of. Where that sits on the illicit commodity curve today ($/kg): $2,400 Cannabis flower $3,165 GB300 NVL72 $3,500 Fentanyl $28,000 Cocaine $65,000 Heroin $138,000 Gold So today NVIDIA ships a product that is denser in value than weed, roughly fentanyl-grade, and still an order of magnitude short of cocaine. For now... GB200 NVL72 was $4M, GB300 is ~$5M, and Vera Rubin NVL72 is already quoted at up to $8.8M at essentially the same rack mass. Rack weight is constrained by the datacenter floor loading; rack price is constrained by nothing. That's ~1.75x $/kg per generation with HBM, CoWoS and power all supply-limited through the decade. Cocaine: Colombia's new president has pledged a hard-line security crackdown with $1B in US aid behind it Every prior crackdown has consolidated the industry into fewer, better-capitalized operators with better logistics and higher yields per hectare. Consolidation is deflationary. We model US wholesale drifting from ~$28k/kg toward ~$12k/kg by 2030 as the supply chain professionalizes. **Crossover: 2030.** Our extrapolated NVL72-class rack hits ~$17k/kg while wholesale cocaine falls to ~$12k/kg. At that point the rational cartel pivots to smuggling racks, except a rack has a fixed 500+ kW power draw, a 3,300 lb forklift requirement and an export-control regime that is actually enforced. Cocaine has none of these problems, which is why it will remain the superior product for anyone without a substation. Key risks to the thesis: US retail coke is still ~$60–200/g, i.e. $60k–200k/kg so the crossover only holds at wholesale. Retail compute (H100 hour on a neocloud) is also marked up, so we consider this an apples-to-apples wholesale comparison. Full model available to Coke Research subscribers.

OPENAI SAYS ITS JALAPENO CHIP BEAT NVIDIA GB300 IN INFERENCE TESTS OpenAI says its new Broadcom-built Jalapeno chip outperformed NVIDIA $NVDA GB300 on both throughput per watt and response latency in testing. Jalapeno runs at roughly 700W and is expected to begin powering OpenAI models later this year, targeting lower-cost and faster inference. Important caveat: it was not tested against NVIDIA’s newer Vera Rubin platform and is not designed for AI training. OpenAI says a second-generation chip is already nearing tape-out, while work on a third generation has begun.

A quick peek at our NVIDIA GB300 NVL72 deployment process. NVIDIA #GB300# NVL72 brings next-generation rack-scale AI performance to the era of reasoning, with ultra-dense compute optimized for large-scale training and high-throughput inference. More to come. Stay tuned👐! #GPU# #AIInfrastructure# #neocloud#

Mizuho on $CRDO: "very positive $MU / CRDO": "Our checks indicate GB300 ramps could continue, significantly stronger than previously expected, into 2H26E and into 2027E - Very Positive MU-CRDO. With GB300 at ~$4M/rack and 145KW, tight memory and land/power/shell (LPS) we see a fat BW tail driving higher rack shipments into 2027E vs VR at ~$8M/rack and ~200KW/800V-400V dual-rail racks. We believe this is consistent with $NVDA recent comments "August Blackwell shipments up 27% m/m" and 2027E GPU shipments up 2x y/y.The upside for connector suppliers comes from increased number of racks/cables, offsetting the downside of lower ASP at 100G/lane vs 200G/lane. For NVDA, we est. ~10% of OctQ units (20% of revenue) on VR/1.6T."

🚀 #NVIDIA’s# rack-scale AI servers are surging: GB300, VR200, and VR300 NVL72 systems are projected to top US$710B in combined output value in 2027, up 214% YoY.💡#TrendForce# on how this could boost ODM, power, and cooling suppliers: 🔗