0xNeoArch

563 posts

0xNeoArch banner
0xNeoArch

0xNeoArch

@0xNeoArch

Software engineer • AI orchestration • Agentic AI SDLC frameworks • Local AI & hardware • Infra • DevOps • Trading • Biohacking Sharing what works

Beigetreten Eylül 2024
3.9K Folgt208 Follower
0xNeoArch
0xNeoArch@0xNeoArch·
GPT-5.6 rumors are everywhere 👀 If OpenAI really ships: • 1.5M token context • Better agent workflows • Lower pricing then the biggest story won't be the model itself. It'll be what autonomous AI agents can build when context windows become effectively "whole-project" sized. June could get interesting. 🚀
English
0
0
0
7
0xNeoArch retweetet
Koda
Koda@wadezone·
4步教你复活 claude fable 5! 1.下载 fable 5的系统提示词 github.com/elder-plinius/… 2.放到你Claude code 项目文件夹 3. 使用启动命令 claude --dangerously-skip-permissions --system-prompt-file CLAUDE-FABLE-5.md 4.模型切换到 opus 4.8 Max 只需要这4步 你会发现 fable 5 又回来了!
中文
134
323
2.2K
272.5K
0xNeoArch retweetet
𝖉𝖊𝖒𝖎
𝖉𝖊𝖒𝖎@demi_hl·
If you have: Hermes Agent Claude Code & Codex Handoffs Obsidian + QMD Memory System Run Agentic Loops Fleet Tailscale Mesh Cron Jobs + Kanban Board Agentic Workflows Congrats you are the top 1% of the AI god stack
English
149
262
3.8K
218.7K
0xNeoArch retweetet
0xSero
0xSero@0xSero·
Sglang is my #1 inference engine. - most active community - most models supported - best performance for 6000s - good GitHub issues I love you sglang ❤️
mr-r0b0t@mr_r0b0t

Don’t miss this @sgl_project update! v0.5.13 brings SM120 DSV4 Optimizations, StepFun 3.7 Flash support, Nemotron Ultra support, updated speculative decoding, diffusion model support, and more! 😍

English
15
7
81
10K
0xNeoArch retweetet
0xSero
0xSero@0xSero·
If you have the NVMe Go download as many models as you think you might ever want. Now, go on Huggingface. They’re coming for open models next.
0xSero tweet media
English
125
89
1.3K
84.5K
0xNeoArch retweetet
Ivan Fioravanti ᯅ
Ivan Fioravanti ᯅ@ivanfioravanti·
Code2LoRA seems an incredibly interesting idea. Qwen2.5-Coder-1.5B is not the most powerful LLM around, but it's enough to validate the concept. Instead of stuffing repository context into the prompt at every query, distill it into a LoRA adapter. One forward pass over the repo snapshot, one adapter, zero extra inference tokens. For evolving codebases, a single layer GRU tracks commit history on top of that snapshot. Each git diff updates the hidden state in <10ms. You get a fresh adapter at every commit without need for a full retraining. Great job Liliana! I bet this will lead to something cool in the near future 🙌
Liliana Hotsko@liliana_hotsko

How do you give a code LLM knowledge of an entire repository without paying for it at every single query? We introduce Code2LoRA: a hypernetwork that turns a repository into its own LoRA adapter. Repo knowledge baked into weights → zero inference-time token overhead.

English
11
33
286
24.6K
Anthropic
Anthropic@AnthropicAI·
The US government, citing national security authorities, has issued an export control directive to suspend all access to Fable 5 and Mythos 5 by any foreign national, whether inside or outside the United States, including foreign national Anthropic employees. The net effect of this order is that we must abruptly disable Fable 5 and Mythos 5 for all our customers to ensure compliance. Access to all other Claude models is not affected. We apologize for this disruption to our customers. We believe this is a misunderstanding and are working to restore access as soon as possible. Read our full statement: anthropic.com/news/fable-myt…
English
12.5K
25.7K
87.8K
89M
0xNeoArch
0xNeoArch@0xNeoArch·
@0xSero awesome, will try it! thank you!
English
0
0
0
34
0xSero
0xSero@0xSero·
From running your local model to droid, pi, opencode, and hermes in 1 click.
English
9
6
119
8K
0xNeoArch
0xNeoArch@0xNeoArch·
@vu3dtu very cool! i want to help with testing. will the Pro version of the ring work, too?
English
2
0
1
286
0xNeoArch retweetet
Saksham
Saksham@vu3dtu·
I loved the idea behind the Google Fitbit Air: an LLM wrapped around your health data, daily briefs, and a coach you can ask questions. But there app is really terrible, it's expensive $100 band plus $10/mo, and Google getting a constant stream of your heart rate, sleep, and other private data. Whoop is worse, with a subscription that runs up to $360 a year. So I bought a $7 generic Chinese smart ring from Temu, reverse engineered its BLE protocol, and built an app around it. Introducing PulseLoop: no subscription, open-source iOS app. Your health data stays on your phone, paired with an AI coach that reads your real ring data, draws charts, and remembers context. Free, bring your own API keys. Demo and code below.
English
26
16
337
88.8K
0xNeoArch retweetet
0xSero
0xSero@0xSero·
We're all sleeping on Step-3.7-Flash. It's phenomenal.
0xSero tweet media
English
51
19
580
40.9K
0xNeoArch retweetet
Ahmad
Ahmad@TheAhmadOsman·
When all these closed labs decide it's time to rug pull everyone, you all are gonna regret not Buying a GPU Owning your compute allows you to be in control, even if partially Not too late yet
English
42
14
234
8.4K
0xNeoArch
0xNeoArch@0xNeoArch·
Fable 5 dropped this morning. The model from Claude, which some have heard as Mythos, was TOO powerful to be released to the public. Well they did release a version of it. I cancelled what I was doing. Built 3 things in real time. One prompt each. What it did made me lose MY MIND. I've tested every model UNDER THE SUN since the early GPTs. Nothing, not 4.8, not GPT-5.5, none of them, NOTHING comes EVEN CLOSE!!! I have seen the future.
English
0
0
0
5