SemiPulse | AI-Powered Semiconductor Supply Chain Intelligence & Market Signals

AI enthusiast adds Nvidia Tesla V100 as loud as a lawnmower to gaming PC for $266 — 32GB of VRAM rig can run 27 billion parameter model at 32 tokens per second - Tom's Hardware

0.82

news.google.com 2026-07-26

Google reportedly developing 'Frozen v2' chip with Gemini's architecture etched into the silicon — engineers project 6 to 10 times more tokens per watt than latest TPUs

0.92

tomshardware.com 2026-07-21

According to a report by The Information, Google is reportedly developing a server chip codenamed 'Frozen v2,' which would integrate parts of the Gemini model's architecture directly into the silicon. This design aims to significantly boost the number of tokens processed per unit of energy, potentia

Google AI chip Gemini model TPU chip architecture AI inference silicon integration energy efficiency

Why Nvidia’s Next AI Battle Is About Tokens Per Watt (NASDAQ:NVDA) - Seeking Alpha

0.95

seekingalpha.com 2026-07-16

Palantir CEO Alex Karp claims AI companies are stealing customers' data while charging them for unproductive tokens

0.78

tomshardware.com 2026-07-03

Palantir CEO Alex Karp delivered a blistering critique of leading AI companies like OpenAI and Anthropic during a CNBC interview, accusing them of stealing customer data while charging for unproductive tokens. He argued that these firms are exploiting customer data to improve their models while simu

AI data security Palantir NVIDIA OpenAI Anthropic AI tokens enterprise data privacy data sovereignty

Nvidia offers to take a cut of AI cloud revenue on top of hardware sales in new optional financing vehicle — trades tokens for revenue cut - Tom's Hardware

0.92

www.tomshardware.com 2026-07-02

NVIDIA has introduced a new business model enabling it to generate revenue twice from the same hardware: once through initial hardware sales and again via a percentage of the ongoing revenue generated by its partners' cloud services. This 'revenue-sharing and credit-support model' aims to provide co

NVIDIA AI Cloud GPU Cloud Computing Semiconductor Business Model Revenue Sharing Data Center

US AI startup Tensordyne claims 3nm Napier chip outperforms NVIDIA Blackwell by 13x in tokens per second - TweakTown

0.92

www.tweaktown.com 2026-06-16

US AI startup Tensordyne has announced the successful tape-out of its 3nm Napier chip, claiming a 13x performance advantage over NVIDIA's Blackwell in tokens per second and a 17x improvement in tokens per watt. The chip, built on TSMC's 3nm process and featuring 138 billion transistors, delivers 2.1

AI chip Semiconductor Artificial Intelligence NVIDIA Tensordyne 3nm process Chip performance Computing power

Tensordyne’s 3nm Napier AI Chip Promises 13x Higher Token Throughput Than Blackwell & Blazes Past Rubin With 1000 Tokens/s In Multi-Trillion Parameter Models - Wccftech

0.92

wccftech.com 2026-06-16

Tensordyne, a US-based AI company, has announced the successful tape-out of its Napier AI chip, which claims to outperform NVIDIA's Blackwell and Rubin chips in token throughput and energy efficiency. Built on TSMC's 3nm process, the Napier chip leverages novel logarithmic AI math, tightly integrate

AI chip Semiconductor 3nm process NVIDIA Tensordyne AI inference Chip performance Compute throughput

Nvidia's Jensen Huang Says AI Has Crossed A Critical Threshold: 'Tokens Are Now Profitable' - Benzinga

0.85

www.benzinga.com 2026-05-26

SkyeChip aims to finalise plans for chip design tokens by year end - The Edge Malaysia

0.7

theedgemalaysia.com 2026-05-20

Cerebras: Deeply Integrated with OpenAI, Reshaping AI Chip Market Expectations with 'Fast Tokens' - 富途牛牛

0.92

news.futunn.com 2026-05-15

Cerebras' deep collaboration with OpenAI is reshaping the AI chip market landscape, reflecting the rapidly growing demand for high-performance computing in AI applications. This partnership signifies a shift from general-purpose to specialized AI computing, driving the semiconductor industry toward

AI chips Semiconductor Artificial Intelligence Chip design Technology partnership Market expectations Computing power Deep learning

NEWS: Three chip design firms get access to Arm's tokens - The Edge Malaysia

0.8

theedgemalaysia.com 2026-05-11

Streaming Tokens and Tools: Multi-Turn Agentic Harness Support in NVIDIA Dynamo - NVIDIA Developer

0.85

developer.nvidia.com 2026-05-09

Semiconductor News & Analysis Feed