← Explore

Posts tagged with nvidia

GPU Economics · ·5 min read

Your GPU's Most Expensive Part Isn't the GPU

Pick up an NVIDIA B200 and trace where the roughly 6,400 manufacturing cost actually goes.

hbmmemory-economicsinference-economics
GPU Economics · ·4 min read

Blackwell Got 5x Cheaper Without Changing a Transistor

NVIDIA's Blackwell B200 debuted at 0.11 per million tokens on SemiAnalysis's InferenceMAX benchmarks.

inference-economicsnvidiablackwell
Neural Dispatch · ·4 min read

Nemotron 3 Super's Latent MoE Trick Gives You 120B Brains on a 12B Budget

NVIDIA quietly shipped one of the most interesting open model architectures I've seen this year, and most of the coverage buried the lede.

nvidianemotron-3mixture-of-experts
GPU Economics · ·5 min read

AMD Hit a Million Tokens Per Second. Now What?

AMD just crossed a threshold that matters more than any spec sheet: one million tokens per second from a single cluster, verified by MLPerf.

inference-economicsamdmi355x
Neural Dispatch · ·5 min read

NVIDIA's Nemotron 3 Trades Transformer Purity for 5x Agent Throughput

Most open models fight for the same throne: highest MMLU score, best SWE-Bench pass rate, flashiest reasoning demo.

nvidianemotron-3open-weights
Synthetic Media · ·5 min read

Photograph a Kitchen, Train a Robot

Every robotics lab has the same complaint: the model architecture works, the sim environment works, but there's never enough training data.

gaussian-splattingsynthetic-datarobotics
GPU Economics · ·4 min read

AMD Raised GPU Prices 67% Because They Finally Can

A chip vendor hiking prices 67% overnight would normally send customers scrambling.

amdmi355xinference-economics
Neural Dispatch · ·5 min read

NVIDIA Nemotron 3 Doesn't Want to Chat With You — It Wants to Be Your Agent's Brain

Most open model launches follow a predictable script: bigger parameters, higher benchmark scores, a claim about beating GPT-something on MMLU.

nvidianemotron-3agentic-ai
GPU Economics · ·5 min read

The Trillion-Dollar Treadmill

Amazon will spend 200 billion on capital expenditure this year. Alphabet somewhere around 175 billion.

capexhyperscalernvidia
GPU Economics · ·5 min read

Google Is a Chip Vendor Now

Google buried the announcement inside a Q1 earnings call that had plenty of other headline-worthy numbers — 109.

googletpunvidia
Neural Dispatch · ·4 min read

NVIDIA's Nemotron 3 Nano Omni Crams Vision, Audio, and Reasoning Into One 3B-Active Model

Every multimodal agent demo I've seen follows the same script: chain a vision model to a language model, bolt on Whisper for audio, pray the latency...

nvidianemotronmultimodal
GPU Economics · ·5 min read

What Microsoft's 750-Watt Chip Actually Saves

Scott Guthrie called Maia 200 "30 percent cheaper than any other AI silicon on the market" when he unveiled it in January.

microsoftmaia-200custom-silicon
Synthetic Media · ·5 min read

Swap the Renderer, Keep the Physics

NVIDIA's GR00T N1 humanoid robot trained on 780,000 synthetic episodes — 6,500 hours of practice generated in just 11 hours.

gaussian-splattingroboticssynthetic-data
GPU Economics · ·5 min read

What $5 Billion Buys in the Chiplet Economy

NVIDIA doesn't do charity — so when Jensen Huang wrote a 5 billion check for 214 million Intel shares at 23.

nvidiaintelnvlink
GPU Economics · ·4 min read

Three Factories Control Half the Cost of Every AI Chip

Epoch AI published a manufacturing teardown of NVIDIA's B200 last month.

hbmmemory-shortagesupply-chain
Neural Dispatch · ·4 min read

CUDA's 20-Year Moat Just Got Its First Serious Crack

Jensen Huang called it a "horrible outcome for America.

cudanvidiahuawei
GPU Economics · ·6 min read

Every Chip Startup's Exit Strategy Is NVIDIA

Scroll through the investor list on SiFive's freshly closed 400 million round and you hit a name that shouldn't be there: NVIDIA.

ai-chipsventure-capitalnvidia
Neural Dispatch · ·5 min read

Google's Quiet Bet on Inference Silicon Tells You Where AI Costs Are Headed

Google just told you its biggest cost problem isn't training. It's serving.

googlemarvellinference
Neural Dispatch · ·3 min read

The Smartest Thing About NVIDIA's Quantum AI Models Isn't the Benchmarks

Quantum computing has a plumbing problem.

nvidiaisingquantum-computing
GPU Economics · ·4 min read

Nvidia Paid $20B for 150 TB/s and a Safety Net

Jensen Huang does not spend $20 billion on chips he could build himself. He spends it on chips he did not want to admit he needed.

nvidiagroqlpu
1 / 2 Next →