𓅃

502 posts

𓅃 banner
𓅃

𓅃

@unawarelist

signal in the noise eyes open mouth shut cloud is someone elses computer 💻 low-iq techsupport | part time agentic tinkerer

🛳 Katılım Şubat 2026
347 Takip Edilen28 Takipçiler
𓅃
𓅃@unawarelist·
all broskis were like "ahhh GLM didn't open source 5.1 aaaaaaa"
Z.ai@Zai_org

Introducing GLM-5.1: The Next Level of Open Source - Top-Tier Performance: #1 in open source and #3 globally across SWE-Bench Pro, Terminal-Bench, and NL2Repo. - Built for Long-Horizon Tasks: Runs autonomously for 8 hours, refining strategies through thousands of iterations. Blog: z.ai/blog/glm-5.1 Weights: huggingface.co/zai-org/GLM-5.1 API: docs.z.ai/guides/llm/glm… Coding Plan: z.ai/subscribe Coming to chat.z.ai in the next few days.

English
0
0
0
22
atlas
atlas@creatine_cycle·
having a gf is insane because it's literally unlimited chat with no token spend
English
284
960
24.6K
1.1M
𓅃
𓅃@unawarelist·
also I'll be asking my agent the same question ☝️⚕️
English
0
0
0
4
𓅃
𓅃@unawarelist·
for folks who use openrouter - a very direct question: is it cost effective? does it really achieve good results for choosing the right model for the right task? (Assuming it's a feature)
English
1
0
0
12
𓅃
𓅃@unawarelist·
@rsuyoy Suggestion: base-app is free bill separately for inference + smart categorization
English
1
0
2
57
𓅃
𓅃@unawarelist·
@rsuyoy it is justified to be pessimistic considering how people are in genera but, giving people the benefit of the doubt (positive) feels warmer though
English
0
0
1
9
Yousr
Yousr@rsuyoy·
@unawarelist We can only hope, tho I am personally much pessimistic than you slop just works, people fall for it easily
English
1
0
1
51
𓅃
𓅃@unawarelist·
I have started to feel that contributing is always more satisfying and wholesome than trying keep up with everything and consuming. Almost as if any form of non-deep consumption doesn't hit right. Very similar to playing a game without cheat codes and in hard-mode, versus playing a game with cheats on, cutting corners and just making "progress". While there is a short term joy in making some progress, the quality of progress starts to accumulate, a depth-debt, which on hindsight doesn't give satisfaction.
English
0
0
0
5
𓅃
𓅃@unawarelist·
finna take the slow pill and have my agent teach me shi + finna do manual work cuz its more rewarding than pressing the button 👍
English
0
0
0
2
𓅃 retweetledi
hopecore
hopecore@dailyhopecores·
hopecore tweet media
ZXX
5
1.6K
8.1K
75.9K
𓅃
𓅃@unawarelist·
more shots fired on the bigclaw vs hermes today huh 🫢👀
English
0
0
0
7
𓅃
𓅃@unawarelist·
@rsuyoy Looking forward to more of your work, you got a long way to go post-uni.
English
0
0
1
12
Yousr
Yousr@rsuyoy·
5 things I’m particularly proud of in my app. An unscripted, spontaneous live demo from my uni dorm :)
English
51
18
525
47.2K
𓅃
𓅃@unawarelist·
the large chunk of tech worker population is still under the impression that AI = ChatGPT phone + webapp and they haven't caught up to MCPs yet
English
0
0
0
16
𓅃
𓅃@unawarelist·
@trendingbots @Teknium @sudoingX @NousResearch More APIs for research, self host easy tools like firecrawl, searxng etc Look into good skills, and get em for your agent And most importantly : just talk to your agent
English
1
0
7
602
𓅃
𓅃@unawarelist·
@rsuyoy SlopFluencer SlopBait
Dansk
1
0
1
35
𓅃
𓅃@unawarelist·
vibecoding leads inevitably to bug driven development you just patch things, and then patch things, and then patch things, and then patch things if something is not doing what it is supposed to be doing
Stone Tao@Stone_Tao

genuine question. how do you debug code and ensure good quality when coding models spit out 1000s of lines i still cannot feel comfortable not understanding what every generated line does, reducing the productivity gains coding models should be giving me

English
0
0
1
39