RSS Reader

submitted 14 days ago * by Wrong-Historian

Here is the thing, the expert layers run amazing on CPU ( ~~~17T/s~~ 25T/s on a 14900K) and you can force that with this new llama-cpp option: --cpu-moe .

You can offload just the attention layers to GPU (requiring about 5 to 8GB of VRAM) for fast prefill.

KV cache for the sequence
Attention weights & activations
Routing tables
LayerNorms and other “

Open Source LLM

OpenAI Rolls Out GPT-5 To All ChatGPT Users As Sam Altman Hails It The Best Model Ever For Writing, Coding, And Human-Level Reasoning

wccftech | Aug. 7, 2025, 1:57 p.m. | Original

After a year of building up its hype, OpenAI has officially its most advanced AI model yet, the GPT-5. It is being made available to all ChatGPT users and developers. CEO Sam Altman compared the debut of the new model to the moment the first iPhone came out with the Retina display - an experience that, once you indulge in it, you can never go back on. The next-generation language model brings some major improvements in reasoning, accuracy, and speed, and brings interactions that are closer

Open Source LLM

OpenAI intros two open-weight language models that can run on consumer GPUs — optimized to run on devices with just 16GB of memory

Tom's Hardware | Aug. 7, 2025, 1:25 p.m. | Original

gpt-oss-120b and 20b are the first open-model LMs since GPT-3

Open Source LLM

OpenAI officially announces GPT-5, its next major upgrade to ChatGPT

9to5mac | Aug. 7, 2025, 1:11 p.m. | Original

OpenAI has officially announced GPT-5, its latest frontier model to power ChatGPT. The announcement follows the release of two open weight models , one of which can run on the Mac , and changes to encourage healthier ChatGPT use this week.

Here’s what OpenAI has to say about GPT-5:

Today we’re introducing , OpenAI’s smartest, fastest, most useful model yet, and a major step towards placing intelligence at the center of every business. GPT‑5 unites and exceeds OpenAI’s

Open Source LLM

OpenAI releases GPT-5, a potential barometer for whether artificial intelligence hype is justified

Citynews | Aug. 7, 2025, 1:08 p.m. | Original

SAN FRANCISCO (AP) — OpenAI on Thursday released the fifth generation of the artificial intelligence technology that powers ChatGPT, a product update that’s being closely watched as a measure of whether generative AI is advancing rapidly or hitting a plateau.

GPT-5 arrives more than two years after the March 2023 release of GPT-4, bookending a period of intense commercial investment, hype and worry over AI’s capabilities.

In anticipation, rival Anthropic released the latest version of its own chatbot, Claude,

Open Source LLM

OpenAI launches GPT-5 as the AI industry seeks a return on investment - Reuters

Reuters | Aug. 7, 2025, 1:02 p.m. | Original

OpenAI launches GPT-5 as the AI industry seeks a return on investment Reuters

Open Source LLM

Duolingo surges as AI-led growth, forecast raise boost investor confidence - Reuters

Reuters | Aug. 7, 2025, 8:03 a.m. | Original

Duolingo surges as AI-led growth, forecast raise boost investor confidence Reuters

Open Source LLM

(PR) SuperX Launches New All-in-One Multi-Model Server Series

TechPowerUp News | Aug. 7, 2025, 7:58 a.m. | Original

Super X AI Technology Limited today announced the official launch of its latest All-in-One Multi-Model Servers ("MMS"). As the first enterprise-grade AI infrastructure to support the dynamic collaboration of multiple models by SuperX, this MMS is centered on being out-of-the-box ready, multimodel fused, and deeply integrated into application scenarios. It provides enterprises with a secure and efficient full-stack AI solutions, with customized specifications for enterprises of various scales, marking a new era where

Open Source LLM

OpenAI poised to release GPT-5 model days after launching 2 new open systems

SCMP | Aug. 7, 2025, 6 a.m. | Original

VID

GPT-5 marks OpenAI’s latest update to the technology behind generative AI chatbot ChatGPT. Photo: Shutterstock

Published: 6:00pm, 7 Aug 2025

San Francisco-based OpenAI hinted as much in an all-caps post on X that read: “LIVE5TREAM THURSDAY 10 AM PT.” While there appeared to be a typo in the first word of that announcement, many of those who posted comments got the hint that the misplaced “5” could be about GPT-5.

In a subsequent post on his personal X account, CEO Sam Altman did not address the speculation on

Open Source LLM

OpenAI 預告 GPT-5 即將登場　或同步推出輕量版 GPT-5 mini、nano

UNWIRE.HK | Aug. 7, 2025, 4:22 a.m. | Original

OpenAI 預告將於香港時間 8 月 8 日凌晨 1 點進行一場直播活動，極有可能正式發表先前傳聞的全新大型語言模型 GPT-5。雖然官方目前仍未明說直播主題，但從預告內容來看，英文「LIVESTREAM」中刻意將「S」改為數字「5」，暗示接下來

120B runs awesome on just 8GB VRAM!

OpenAI Rolls Out GPT-5 To All ChatGPT Users As Sam Altman Hails It The Best Model Ever For Writing, Coding, And Human-Level Reasoning

OpenAI intros two open-weight language models that can run on consumer GPUs — optimized to run on devices with just 16GB of memory

OpenAI officially announces GPT-5, its next major upgrade to ChatGPT

OpenAI releases GPT-5, a potential barometer for whether artificial intelligence hype is justified

OpenAI launches GPT-5 as the AI industry seeks a return on investment - Reuters

Duolingo surges as AI-led growth, forecast raise boost investor confidence - Reuters

(PR) SuperX Launches New All-in-One Multi-Model Server Series

OpenAI poised to release GPT-5 model days after launching 2 new open systems

OpenAI 預告 GPT-5 即將登場　或同步推出輕量版 GPT-5 mini、nano

Menu