Articles Tagged “Custom Silicon”
9 articles found
Qualcomm Amazon Deal Ties a $4B Warrant to Chip Orders
Qualcomm shares rose 3% after Amazon took warrants on 25 million shares at $161.26, with vesting tied to purchases counted up to $60 billion by 2036.
Broadcom Q3 Earnings Test a $16B AI Revenue Guidance
Broadcom reports fiscal Q3 on September 2 against its own guidance of $29.4B in revenue and $16B in AI semiconductor sales, up over 200% YoY.
OpenAI Jalapeño Benchmarks: 1.9x Work per Kilowatt
OpenAI published the first Jalapeño inference benchmarks: a 700W part claiming up to 1.9x throughput per kilowatt and 3.6x lower end-to-end latency.
Meta MTIA 400 Puts 9.4TB/s HBM3e Behind FP4 Inference
Meta detailed MTIA 400 at Hot Chips 2026: eight HBM3e stacks, 9.4TB/s of bandwidth, hardware FP4, and scale-up domains reaching 72 accelerators.
Marvell Google TPU Deal Comes With a $12.2B Warrant
Marvell expanded its custom silicon work for Google's TPU ecosystem and issued a warrant for up to 58.97 million shares at $206.58, worth $12.2B.
Etched Raises $700M at a $21B AI Chip Valuation
Etched closed a $700M Series D led by Jane Street at a $21 billion valuation, doubling in just a month as inference chip orders pass $1 billion.
OpenAI and Broadcom's "Jalapeño" Chip: Custom Silicon for Inference
OpenAI's first custom chip, Jalapeño, is an inference-focused ASIC co-built with Broadcom, promising better performance-per-watt for large language models.
OpenAI and Broadcom Unveil Jalapeño, a Custom AI Inference Chip
OpenAI and Broadcom unveiled Jalapeño on June 24, 2026 — OpenAI's first custom AI inference chip, promising better performance-per-watt and major cost savings for everyday AI.
This Finnish Startup Burned Llama 3.1 Directly Into Silicon — And It Runs at 17,000 Tokens Per Second
Taalas HC1 eliminates the memory bottleneck entirely by hardwiring an LLM into the chip itself, achieving 10x the speed of an NVIDIA H100.
