Aslan
@avatsaev
Followers
609
Following
2K
Media
606
Statuses
5K
Software engineer working with web technologies, data, and AI.
France
Joined February 2011
Engineer's Guide to running Local LLMs with #llamacpp on Ubuntu, @Alibaba_Qwen Coder 30B running locally along with QwenCode in your terminal https://t.co/DZC9C25YUO
dev.to
Introduction In this write up I will share my local AI setup on Ubuntu that I use for my...
0
0
0
the Qwen bros COOKED with this one Qwen3 Next 80B-A3B > DeepSeek V3.1-level intelligence > at less than half the size extremely smart & efficient > 256k context > instruct & thinking variants > 3B active params per token (3.8%) run this at home on 8x RTX 3090s Buy a GPU ;)
16
12
166
Qwen3-30B-A3B-Instruct repetitions problem running in llamacpp, my sampling params: `--ctx-size 4000 --temp 0.7 --top-p 0.8 --top-k 20 --min-p 0 --repeat-penalty 1.05 --presence-penalty 1.05 --flash-attn --mlock` did anyone manage to fix this? im running a quantized version in
1
0
0
Locally running new #qwen3 instruct model (via #llamacpp), in a locally running chat UI, with locally running #MCP servers and locally hosted search engine, the speeds here are insane 🚀 The new @Alibaba_Qwen A3B Instruct is a very impressive model, congrats to the team!
0
0
4
Large Enough | Mistral AI | Frontier AI in your hands
mistral.ai
Today, we are announcing Mistral Large 2, the new generation of our flagship model. Compared to its predecessor, Mistral Large 2 is significantly more capable in code generation, mathematics, and...
0
0
0
The xz fiasco has shown how a dependence on unpaid volunteers can cause major problems. Trillion dollar corporations expect free and urgent support from volunteers. @Microsoft @MicrosoftTeams posted on a bug tracker full of volunteers that their issue is "high priority"
192
3K
15K
Yi model paper - A comprehensive guide on training and finetuning robust models. In an age where detailed methodologies are often withheld, this paper offers deep insights into data collection, filtration, mixing, and the processing of SFT data https://t.co/n4piyWXoIX
#ai #ml
arxiv.org
We introduce the Yi model family, a series of language and multimodal models that demonstrate strong multi-dimensional capabilities. The Yi model family is based on 6B and 34B pretrained language...
0
0
0
Today we’re releasing Code Llama 70B: a new, more performant version of our LLM for code generation — available under the same license as previous Code Llama models. Download the models ➡️ https://t.co/fa7Su5XWDC • CodeLlama-70B • CodeLlama-70B-Python • CodeLlama-70B-Instruct
163
1K
5K
A Next.js app requires dozens of config files — next.config.js, eslintrc.json, tsconfig.json, package.json, postcss.config.js, tailwind.config.js, and more. How did we get here? How do we avoid it? https://t.co/3UogVY7OhC
deno.com
Why a Next.js project has over 30 configuration files and what we can do to avoid it.
17
58
347
