Open Source LLM

(PR) NVIDIA Announces Generative AI Models and NIM Microservices for OpenUSD Language, Geometry, Physics and Materials

NVIDIA today announced major advancements to Universal Scene Description, or OpenUSD, that will expand adoption of the universal 3D data interchange framework to robotics, industrial design and engineering, and accelerate developers' abilities to build highly accurate virtual worlds for the next evolution of AI.

Through new OpenUSD-based generative AI and NVIDIA-accelerated development frameworks built on the NVIDIA Omniverse platform, more industries can now develop applications for visualizing industrial design and engineering projects, and for simulating environments to build the next wave of physical AI and robots.

Open Source LLM

libX11 1.8.10 Brings Memory Safety Fixes

Alan Coopersmith of Oracle -- thanks to his work on Solaris and maintaining the X11 support -- continues to be one of the few developers left managing new X.Org software component releases. This weekend Coopersmith released libX11 1.8.10 as the newest version of this client-side library for the core X11 protocol.

The libX11 1.8.10 release has various fixes and improvements that have accumulated over the past several months. Among the work are some

Open Source LLM

Chinese AI built off open-source code matches American tech in chatbot benchmark tests

Abstract stream of information (Image credit: Shutterstock)

As per The New York Times , Chinese firm 01.AI’s Yi-Large Global SOTA LLM, based on open-source technology, scored nearly as well as its American counterparts, like GPT-4o and Llama-3. Another AI tool called Kling, a video generator, is also publicly available in China. It works similarly to Open AI’s Sora, but unlike the latter, which is still limited access, Kling is already widely

Continue reading...

Open Source LLM

LLVM 19.1-rc1 Compiler Released With More C23 / C++23 & New Intel Extensions

LLVM 19.1-rc1 was released today as the first tagged development snapshot of LLVM 19 that is working its way toward the stable LLVM 19.1 version expected in September.

Earlier this week when the LLVM 19 code was branched I wrote an article outlining all of the new features in LLVM 19 and Clang 19 . See that article for more details but there's a lot in new C and C++ language features, Intel Branch Hint support, new Arm cores added
Continue reading...

Open Source LLM

AI趨勢周報第257期: Mistral AI用Mamba架構打造程式碼生成模型

重點新聞(0719~0725)

Mamba Mistral AI 程式碼

Mistral AI用Mamba架構打造程式碼生成模型

Mistral AI最近開源一款70億參數的程式碼生成模型Codestral Mamba,特別的是,這款模型不以Transformer架構為基礎,而是採用後來出現的Mamba 2架構。Mistral AI解釋,有別於

Continue reading...

Open Source LLM

ollama v0.3 Released With Llama 3.1 Support & Mistral Large 2

For fans of ollama as an open-source means for easily running large language models (LLMs) on your system, ollama v0.3 has been released with support for the newest exciting models.

The ollama v0.3 pre-release that debuted overnight has initial support for Llama 3.1, the newest LLM in the Llama family from Meta/Facebook. Llama 3.1 was introduced earlier this week by Meta as their "most capable" model to date with their 405B version claiming
Continue reading...

Open Source LLM

Mistral AI發表Mistral Large 2,挑戰GTP-4o與Llama 3.1 405B

去年才成立的Mistral AI腳步飛快,於本週三(7/24)推出了具備1,230億個參數,支援12.8萬個Token脈絡長度的 Mistral Large 2 ,此一Mistral AI的旗艦模型宣稱已可比美OpenAI的 GTP-4o 及Meta剛開源的 Llama 3.1 405B

Mistral Large 2支援包括中


Open Source LLM

OpenAI公佈GPT-4o mini微調工具對抗最新Meta Llama 3模型,免費使用2個月

Meta週三發布最新版大語言模型Llama 3.1,來勢洶洶,OpenAI則在宣佈最新小模型GPT-4o mini不到一週後, 再公佈微調工具 ,提供免費使用特定用量2個月。

微調工具 已可由API使用,一開始提供給tier 4和tier 5的用户。OpenAI計畫會逐步推展


Open Source LLM

Meta開源Llama 3.1 405B,強調開源才是AI的未來

Meta週二(7/23) 開源了Llama 3.1 405B ,它具備4,050億個參數,脈絡長度達12.8萬個Token,支援8種語言,這是Meta迄今所開發的最大模型,亦號稱是全球第一個達頂尖水準的開源模型,也是全球最大也最有能力的

Continue reading...

Open Source LLM

AMD ROCm 6.2 Release Appears Imminent For Advancing Open-Source GPU Compute

We appear to be on the heels of the AMD ROCm 6.2 software release for advancing the open-source AMD Radeon/Instinct GPU compute stack with new features.

With all the industry interest around the AMD Instinct MI300 series and beginning to gain more traction in the GPU compute space, there has been renewed interest and attention on the ROCm software stack especially with AMD working as well to bring formal ROCm support to more consumer GPUs. As such, we're very eager for ROCm 6.2
Continue reading...