Новости TSMC
Железо / ЧипыОбновлено: 16.11.2026
C.C. Wei
Hsinchu
Осн. 1987
NYSE:TSM
Об TSMC
Контрактный foundry. Узлы N2/N3 — сущность чипа, не тикер.
Новости TSMC
Все 278 →Show HN: Goethe – turn your computer into an inference engine
Hi HN, I'm Matteo from meshapplied (YC W27), building this with Peter and Janos.We think you can and should own your inference, so we built software that actually supports that. Goethe…
Qwen3-TTS in ComfyUI: Multi-Speaker Dialogue, 3-Second Voice Cloning, and One Sampler You Have Not Seen
Voice generation finishing inside ComfyUI changes the shape of a video pipeline: audio and picture become one graph instead of two tools and an export. This release puts Qwen3-TTS in the…
Jailbroken PS4 runs 9-billion-parameter AI model at up to 43 tokens per second
Garlic LLM runs Alibaba’s Qwen3.5-9B on a jailbroken PlayStation 4’s GPU, fully offline, at 31 to 39 tokens per second.
Working With UHF RFID Tag Data: EPC, TID, RSSI and Duplicate Reads Explained
If you're writing software around UHF RFID for the first time, such as a warehouse app, a gate system or an asset register, the reader will hand you a stream of reads that looks simple but…
Jensen Huang's resignation causes dismay and disbelief in Taiwan, his birthplace and where he is "the people's dad"; an official expects little impact on TSMC (Li Yuan/New York Times)
Li Yuan / New York Times: Jensen Huang's resignation causes dismay and disbelief in Taiwan, his birthplace and where he is “the people's dad”; an official expects little impact on TSMC —…
У нас есть Claude дома — заметка о том, как я делал быстрый и дешёвый локальный LLM
Берём видеокарту RTX 5060 Ti 16 ГБ, 64 ГБ быстрой DDR5 памяти, запускаем небольшую Qwen3.8 27B Q3 во всю vram, с околонулевым контекстом, получаем неплохую скорость, пишем сапёр/змейку — и…
Alibaba’s Qwen4-Max targets AI’s cost and context bottlenecks
Quick take Alibaba says average cost per task has fallen by more than 60%, while Qwen4-Max uses up to roughly 50% fewer tokens than Qwen3.8-Max. The model’s listed API pricing is $1.50 per…
Alibaba выпустила Qwen4-Max в API по токенам: на 25% дешевле Qwen3.8-Max
Qwen4-Max стоит $1,50 и $4,50 за миллион токенов; в независимом индексе Artificial Analysis у неё 43 балла и 38-е место, ниже Kimi K3.1 Pro и Opus 5.5.— Читать дальше «Alibaba выпустила…
Нейросеть на железе со свалки, или Как Bonsai 2 27B проиграла Qwen3 8B
У меня было 8 гигабайт видеопамяти на старой майнинговой NVIDIA P104-100 семнадцатого года, древний i7-3770 родом из двенадцатого, знающий только AVX, ровно 15,5 гигабайта оперативной…
Alibaba выпустили Qwen4-Max
Alibaba представили Qwen4-Max, новую флагманскую модель с контекстом на 1 млн токенов и поддержкой изображений. В тестах команды она расходует до 50% меньше токенов, чем Qwen3.8-Max. Цена…
Qwen4-Max scores 43 on the Artificial Analysis Intelligence Index, beating Qwen3.8 Max at ~67% lower cost per task but trailing Kimi K3.1 Pro at ~11% lower cost (@artificialanlys)
@artificialanlys: Qwen4-Max scores 43 on the Artificial Analysis Intelligence Index, beating Qwen3.8 Max at ~67% lower cost per task but trailing Kimi K3.1 Pro at ~11% lower cost — Alibaba…
Alibaba releases Qwen4-Max, its new flagship 2.4T-parameter MoE model with a 1M-token context window, saying it beats Qwen3.8-Max with up to ~50% fewer tokens (Qwen)
Qwen: Alibaba releases Qwen4-Max, its new flagship 2.4T-parameter MoE model with a 1M-token context window, saying it beats Qwen3.8-Max with up to ~50% fewer tokens — Today, we are…
Qwen4-Max Is Here: A 2.4T-Parameter Teammate That Cuts Your Agent Bill by 62% — With a Catch
Alibaba's Qwen team dropped a new flagship today: Qwen4-Max, built on a brand-new Qwen4 architecture with a 2.4-trillion-parameter backbone, native vision, and a one-million-token context…
Sources: Nvidia appealed to Taiwanese officials after its manager was charged over chip smuggling, and sought advice from TSMC, which declined to get involved (Mackenzie Hawkins/Bloomberg)
Mackenzie Hawkins / Bloomberg: Sources: Nvidia appealed to Taiwanese officials after its manager was charged over chip smuggling, and sought advice from TSMC, which declined to get involved…
На PlayStation 4 запустили ИИ-модель Qwen3.5-9B с 9 млрд параметров
У консоли всего 8 ГБ унифицированной памяти PlayStation 4 вышла в далёком 2013 году, но это не помешало энтузиасту Леонардо Алегрия создать проект, который позволяет запускать на этой…
Python 3 Standard Library Socket Programming Cookbook: TCP, UDP, select.select(), struct Packing, and Checksums
A zero-dependency cookbook for beginners. Everything below uses only the Python 3 standard library (socket, select, struct, hashlib, zlib, json, sys), flat procedural code, no asyncio,…
Ускоряем AI модель на процессоре Qwen3.8-Flash-Next Gyro
Существует много различных проектов и вариантов по ускорению генерации AI. Но в большинстве случаев они используют тюнинг модели, а так же использование внешней видео карты. Но можно ли…