Custom ASIC 3.0: hyperscalers now mix and match suppliers across compute and networking
BenBajarin · x · 2026-09-15
Ben Bajarin's Custom Silicon 3.0 report shows the ASIC market has shifted from design wins to program responsibility: Google and Amazon deepen in-house tooling while mixing compute, I/O, networking and packaging services across multiple suppliers. The report separates compute from networking for finer sourcing combos, and argues GPUs remain the strongest general-purpose choice on TCO as inference workloads keep diversifying.
More from Infra
- NVIDIA: full-stack NIM tuning delivers 2.5x more concurrent users on Nemotron 3 Ultra — NVIDIAAI · 2026-09-15
- How Much Does Local LLM Inference Really Cost? A Dev Added an Electricity Calculator — giveen · 2026-09-15
- Hugging Face Rounds Up Which Open LLMs Are Best for On-Device Inference — NielsRogge · 2026-09-15
- Single Pure-C99 Inference Engine Runs Both BitNet Ternary and GGUF, No Python or CUDA — shifu_legend · 2026-09-15
- Dev weighs ChatGPT subscription via OAuth vs API pricing for a production RAG app — builtforoutput · 2026-09-15
- Cognichip launches ACI Enterprise: one engineer finishes chip front-end design in 10 days — kimmonismus · 2026-09-15