DeepX tapes out 2nd-gen chip claiming 100B-parameter inference under 5W with LPDDR5X PIM
giuliomagnifico · hn · 2026-11-10
Per EE Times, edge-AI chip startup DeepX has taped out its second-generation NPU built on LPDDR5X PIM (processing-in-memory). The company claims it can run inference for models up to 100B parameters at under 5 watts, projecting a 6-7x performance gain over the previous generation — a notable datapoint for on-device LLM deployment power budgets.
More from Infra
- Batch APIs: the 50% LLM discount only pays off after model choice and prompt caching — arpit_bhayani · 2026-11-13
- RTX Pro 6000 bricks mid-Qwen run, owner jokes it "died of a broken heart" over Jensen news — wine_dark · 2026-11-13
- Bittensor's Chutes Handles 256 Concurrent Requests on a Single RTX 5090 at ~18k tok/s — markjeffrey · 2026-11-13
- Google expands TPU v8 lineup: 8tx cuts latency 2x, 8i Max doubles silicon per rack — Beth_Kindig · 2026-11-13
- Horace He to share two mental models for reasoning about inference at SGLang Summit — cHHillee · 2026-11-13
- Used A100 prices fall another 20-30% as sellers rush ahead of a collapse — Bascilian · 2026-11-13