--- format: "aidr-story-markdown/v1" id: "3fcb3a86ea59b36a84336a5b307a65bb0f33011bc20438e95970de963a4d9136" canonical_url: "https://aidr.today/3fcb3a86?lang=en" title: "Nvidia Denies China LPU Shipments by Year End in Reversal of Reported Comeback" lang: "en" requested_lang: "en" available_langs: ["en","vi"] translation_fallback: null fallback_fields: [] published_at: "2026-08-20T20:54:28.000Z" category: "Chips" topics: ["nvidia","chips","inference"] source_urls: ["https://huggingnews.com/ai/nvidia-denies-china-lpu-shipments-by-year-end-in-reversal-of-reported-co-54a07a2e","https://x.com/FirstSquawk/status/2090529186922553833","https://x.com/theinformation/status/2090514290549600387","https://x.com/wallstengine/status/2090484893100994638","https://x.com/Reuters/status/2090529697759146049","https://x.com/financialjuice/status/2090484058811699602","https://x.com/StockMKTNewz/status/2090485916078874748","https://x.com/zerohedge/status/2090485051091726812"] summary: "The chipmaker has no China-specific Language Processing Unit (LPU) on its roadmap and is not conducting sales of the product in the Chinese market. Nvidia stated that reports of small-batch shipments of a custom AI inference chip to Chinese clients by the end of 2026 are incorrect. The Information previously reported that the LPU used Groq-licensed technology and complied with US export rules without reducing computing power. Those reports claimed several Chinese customers had already placed orders for the chip to accelerate chatbot response times alongside graphics processing units. This potential reentry followed comments from CEO Jensen Huang that the company had largely conceded the market to Huawei." --- # Nvidia Denies China LPU Shipments by Year End in Reversal of Reported Comeback > [Open the canonical story]() **Published:** 2026-08-20T20:54:28.000Z **Category:** Chips **Topics:** nvidia, chips, inference ## Summary The chipmaker has no China\-specific Language Processing Unit \(LPU\) on its roadmap and is not conducting sales of the product in the Chinese market\. Nvidia stated that reports of small\-batch shipments of a custom AI inference chip to Chinese clients by the end of 2026 are incorrect\. The Information previously reported that the LPU used Groq\-licensed technology and complied with US export rules without reducing computing power\. Those reports claimed several Chinese customers had already placed orders for the chip to accelerate chatbot response times alongside graphics processing units\. This potential reentry followed comments from CEO Jensen Huang that the company had largely conceded the market to Huawei\. ## Sources - [Story source]() - [Story source]() - [Story source]() - [Supporting source]() - [Supporting source]() - [Story source]() - [Supporting source]() - [Supporting source]()