AI hardware | byteiota

Tag: AI hardware

Split-screen comparison of Alibaba Zhenwu V900 AI chip versus blocked Nvidia accelerator with export restriction barrier

News

Alibaba Zhenwu V900: No FLOPS, 216GB RAM, and Qwen 5 at 10T

Alibaba T-Head's Zhenwu V900 ships 216GB HBM and 500K-chip clusters — but no FLOPS disclosed. ...

By ByteBot

23 hours ago

Conceptual illustration of OpenAI Jalapeño chip with neural network lines converging on a microchip, representing LLM-assisted chip design

News

OpenAI’s Jalapéño Chip: Designed by LLMs, Beats Nvidia

OpenAI taped out the Jalapéño inference chip in 9 months with ~100 engineers, using its ...

By ByteBot

7 days ago

NVIDIA NVLink Fusion connecting custom XPUs and AI accelerators in a data center rack with glowing blue interconnect lines

AI & Development

NVIDIA NVLink Fusion: What the $3.5B MediaTek Deal Really Changes

NVIDIA's $3.5B MediaTek investment is really about NVLink Fusion — the platform that lets custom ...

By ByteBot

September 8, 2026

News

Etched Ships First Transformer ASIC Rack at $21B Valuation

Etched delivered its first Sohu transformer ASIC rack to Jane Street and raised $700M at ...

By ByteBot

August 25, 2026

Stylized HBM memory chip with circuit patterns representing Micron Research Labs AI memory initiative

News

Micron Research Labs Launches With $10B AI Memory Bet

Micron just announced a $10B AI memory research hub backed by NVIDIA, Apple, and the ...

By ByteBot

August 21, 2026

Cerebras CS-4 wafer-scale chip cluster with blue circuit patterns and speed lines representing 30x faster AI inference

News

Cerebras CS-4 Goes 30x Faster: OpenAI Bets Billions

Cerebras CS-4 launches today with 4,400 tokens/second and a 750MW OpenAI deal. Here's what the ...

By ByteBot

August 19, 2026

Split-screen illustration showing Groq LPU chip absorbed into Nvidia GPU infrastructure, representing the company's pivot from chip maker to neocloud provider

News

Groq’s Neocloud Pivot: Nvidia Took Its Soul in 2026

Groq raised $350M at half its peak valuation. Nvidia took its founder, its tech, and ...

By ByteBot

August 18, 2026

Mojo programming language 1.0 release - GPU chip with Python-style design in ByteIota blue

News

Mojo 1.0 Is Here: Python Speed, Rust Safety, AI Hardware

Modular shipped Mojo 1.0 on August 12 — the first stable, production-ready release. Here's what ...

By ByteBot

August 17, 2026

Aerial view of the Terafab semiconductor chip manufacturing facility in Texas with glowing blue circuit traces

News

Tesla and SpaceX’s Terafab: $16.8B Chip Factory

Tesla and SpaceX broke ground on Terafab, a $16.8B semiconductor megafactory in Texas. Here's what ...

By ByteBot

August 8, 2026

News

Samsung zHBM: Memory Stacked on AI Chips — 8x HBM5 Speed

Samsung unveiled zHBM at FMS 2026 — a memory architecture that stacks HBM directly above ...

By ByteBot

August 7, 2026

12 3

Tag: AI hardware

Posts navigation

Categories

Latest Posts