๊ณ ์ ๋ ํธ์
Alioune ๐ธ๐ณ๐ง๐
10.9K posts
Alioune ๐ธ๐ณ๐ง๐
@ali0une
in code i trust ... wind0w$ suxX, GNU/#Linux roxX! check @rastavibesnet #reggae shop #debian #php #sql #geek #Bordeaux #Senegal #Africa https://t.co/bURQId4Nif
Bordeaux FRANCE Dakar SENEGAL ๊ฐ์
์ผ Mart 2010
171 ํ๋ก์146 ํ๋ก์
@zhimin_zhang @ggerganov none, AFAIK speculative decoding uses the same loaded model (-m or --model) Qwen3.6-27B here
English
Alioune ๐ธ๐ณ๐ง๐ ๋ฆฌํธ์ํจ
Alioune ๐ธ๐ณ๐ง๐ ๋ฆฌํธ์ํจ
Let me demonstrate the true power of llama.cpp:
- Running on Mac Studio M2 Ultra (3 years old)
- Gemma 4 26B A4B Q8_0 (full quality)
- Built-in WebUI (ships with llama.cpp)
- MCP support out of the box (web-search, HF, github, etc.)
- Prompt speculative decoding
The result: 300t/s
(realtime video)
English
@nixcraft Would not touch microslop even with a stick ๐
English
I still wonโt use Microslop even if give me weekly million dollar bonus ๐
Zac Bowden@zacbowden
BREAKING: Microsoft just announced several major changes to Windows 11 in an effort to win back user trust and evolve the platform into something people will actually want to use over macOS and Linux! It's a huge announcement that addresses Windows 11's biggest problems today, tackling core fundamental issues such as unreliable system performance, UX consistency, AI bloat and general enshittification. Microsoft has confirmed that this year, it WILL be reducing where ads and Copilot appear throughout the system, including in Start, Widgets, Notepad, Photos, and more! File Explorer and Windows Search will be upgraded with improved performance and capabilities that make finding apps and files much faster and easier. The OS will become lighter with less RAM and system utilization at idle, making it smoother to run on low end hardware with limited memory. These improvements will also benefit high-end PCs too. Windows Update will be improved with more granular controls and the ability to postpone updates for longer, along with reducing how often the OS needs to restart to install an update. Microsoft has also confirmed that it's bringing back fan favourite features such as the ability to move the Taskbar! It's also working to update more areas of the system shell with modern WinUI designs, which should make Windows 11 feel more coherent and complete. There's much more in the announcement, and it honestly all sounds too good to be true. Microsoft really is listening to feedback, and is eager to make Windows the BEST desktop OS on the market. More details including when these changes will arrive in the link! windowscentral.com/microsoft/windโฆ
English
รa fait longtemps que Riss n'est plus drรดle, mais lร il touche le fond.
Rokhaya Diallo@RokhayaDiallo
Dans le droit fil de lโimagerie coloniale, Charlie Hebdo incapable de confronter les idรฉes dโune femme noire sans la rรฉduire ร un corps dansant, exotisรฉ, supposรฉment sauvage. Ce dessin hideux vise ร me rappeler ma place dans la hiรฉrarchie raciale et sexiste. (via @catboychevik)
Franรงais
Alioune ๐ธ๐ณ๐ง๐ ๋ฆฌํธ์ํจ
Alioune ๐ธ๐ณ๐ง๐ ๋ฆฌํธ์ํจ
The new WebUI in combination with the advanced backend capabilities of llama.cpp delivers the ultimate local AI chat experience
It's fast, private, free and open-source
It runs on any hardware - today
Huge thanks to the team at @huggingface for initiating, leading and supporting this work. Enjoy!
github.com/ggml-org/llamaโฆ
English
Alioune ๐ธ๐ณ๐ง๐ ๋ฆฌํธ์ํจ
llama.cpp developers and community came together in a really impressive way to implement Qwen3-VL models. Check out the PRs, itโs so cool to see the collaboration that went into getting this done.
Standard formats like GGUF, combined with mainline llama.cpp support ensures the models you download will work anywhere you choose to run them.
This protects you from getting unwittingly locked into niche providersโ custom implementations that wonโt run outside their platforms.
Qwen@Alibaba_Qwen
๐ย Qwen3-VLย isย nowย availableย onย llama.cpp! Runย thisย powerfulย vision-languageย modelย directlyย onย yourย personalย devicesโfullyย supportedย onย CPU,ย CUDA,ย Metal,ย Vulkan,ย andย otherย backends. Weโveย alsoย releasedย GGUFย weightsย forย allย variantsโfromย 2Bย upย toย 235B.ย Downloadย andย enjoy!ย ๐ ๐ค Hugging Face: huggingface.co/collections/Qwโฆ ๐ค ModelScope: modelscope.cn/collections/Qwโฆ ๐ PR: github.com/ggerganov/llamโฆ
English
Alioune ๐ธ๐ณ๐ง๐ ๋ฆฌํธ์ํจ
๐ย Qwen3-VLย isย nowย availableย onย llama.cpp!
Runย thisย powerfulย vision-languageย modelย directlyย onย yourย personalย devicesโfullyย supportedย onย CPU,ย CUDA,ย Metal,ย Vulkan,ย andย otherย backends.
Weโveย alsoย releasedย GGUFย weightsย forย allย variantsโfromย 2Bย upย toย 235B.ย Downloadย andย enjoy!ย ๐
๐ค Hugging Face: huggingface.co/collections/Qwโฆ
๐ค ModelScope: modelscope.cn/collections/Qwโฆ
๐ PR: github.com/ggerganov/llamโฆ
English
Alioune ๐ธ๐ณ๐ง๐ ๋ฆฌํธ์ํจ
Alioune ๐ธ๐ณ๐ง๐ ๋ฆฌํธ์ํจ
Une guillotine vient dโรชtre amenรฉe au milieu de la manifestation parisienne #18septembre
Franรงais
Alioune ๐ธ๐ณ๐ง๐ ๋ฆฌํธ์ํจ
Alioune ๐ธ๐ณ๐ง๐ ๋ฆฌํธ์ํจ
Alioune ๐ธ๐ณ๐ง๐ ๋ฆฌํธ์ํจ
@rudiservo @ggerganov @nomic_ai Thanks for the forward, i build llama.cpp on a Debian box and use mostly through its api with codium with continue or through scripts. LM Studio aslo.
English
@ali0une @ggerganov @nomic_ai You're welcome.
if you use docker compose you can set it up with openweb-ui with one file.
For code you can use llamaccp add-ons or go with tabbyml.
English
English
English