🫒☀️COLD PRESSED ADVICES ☀️🫒

422 posts

🫒☀️COLD PRESSED ADVICES ☀️🫒 banner
🫒☀️COLD PRESSED ADVICES ☀️🫒

🫒☀️COLD PRESSED ADVICES ☀️🫒

@Helios_Rises

Bro-guidance | Not medical advice.

Katılım Aralık 2023
179 Takip Edilen15 Takipçiler
Sabitlenmiş Tweet
🫒☀️COLD PRESSED ADVICES ☀️🫒
🫒☀️COLD PRESSED ADVICES ☀️🫒@Helios_Rises·
2 month update on my fertility journey. From 0 sperm to 5 million. Still low, but progressing better than expected. I expect to be in range in the next 2 months. It takes 3 months for sperm to reflect changes affected today.
🫒☀️COLD PRESSED ADVICES ☀️🫒 tweet media
English
0
0
0
61
G MAN 55
G MAN 55@G333M4N·
@zerohedge It's far higher than that, the NDIS fraud would push it to 15% all these people working in an industry zero productivity, $50-60 Billion worth of people doing nothing...
English
1
1
16
590
zerohedge
zerohedge@zerohedge·
Australia Jobless rate rose to 4.3% (estimate 4.1%) from 4.1% in January
English
31
36
386
114.4K
Felix Rieseberg
Felix Rieseberg@felixrieseberg·
We're shipping a new feature in Claude Cowork as a research preview that I'm excited about: Dispatch! One persistent conversation with Claude that runs on your computer. Message it from your phone. Come back to finished work. To try it out, download Claude Desktop, then pair your phone.
English
942
1.5K
17.3K
6M
dr. z, esq.
dr. z, esq.@zeynepmyenisey·
The placebo effect is CRAZY. There's a study from 2002 where 180 people with osteoarthritis were split into three groups. One got standard surgery, one group got saline splashed in there and sewn back up, and the third got a totally fake surgery where they were cut open and closed up without doing anything. The people who got the fake surgery had just as much relief in pain and mobility as the people who actually got the surgery. I actually think this kinda gets into expectancy theory and conditioned cellular response and it proves that your cells are a massive, decentralized network of biological sensors and you are a programmable organism
Diana S. Fleischman@sentientist

Guy loses 10 pounds in a week after he thinks he injected himself with a GLP-1

English
86
634
16.1K
1.6M
Macroblock
Macroblock@sainimatic·
For people asking if you'll be able to turn off DLSS5, it's better than that : you won't be able to turn it on because that demo runs on two 5090s
English
65
731
21.6K
409.5K
Old Media
Old Media@oldmedia·
This scene from Smallville kills me every time
English
194
1.1K
29.1K
3.6M
Tan Man
Tan Man@reallytanman·
Toddlers aren’t meant to use diapers Proctor and Gamble propagandized everyone to delay potty training so they could sell more starting in the ‘50s But our baby is 2 years old and we have used maybe 5 diapers in the past 6 months @oliveoilqueen starting potty training at 6 weeks old Easiest way to reduce the cost and waste of diapers is to just train babies early
Wellness Mama@WellnessMama

Every disposable diaper ever made is still in a landfill. Every single one. 400–500 years to break down. A founder just spent 4.5 years working with fungi scientists to change that... starting with baby diapers. New episode with @twinmiki🎙️🍄 - wellnessmama.com/podcast/1037/

English
8
5
139
30.7K
akira
akira@realmcore_·
I don't think there's a single terminal ux that handles agent swarms well With slate, you can literally use Opus 4.6 and GPT 5.4 at the exact same time But making it intuitive took a ton of work So heres a thread on how it works and how to actually use it 🧵
English
54
66
1.1K
134.1K
Moe 💎
Moe 💎@vvsmoe·
These days, whenever I see a long tweet, I skim it first. Gotta make sure I’m not about to waste my time on AI slop
English
1
0
1
49
Shubh Agrawal
Shubh Agrawal@ShubhAgrawal26·
The thing with Claude and Claude code is that it's too eager to build something like, "Here's the document. Here's the tool. Here's the framework. Here's a spec." Dude, I need you to calm down. Let me discuss. Let me tell you my opinions. Tell me what you think? Take it easy, man
English
247
58
2.1K
158.5K
Aus Integrity
Aus Integrity@QBCCIntegrity·
The Reserve Bank of Australia will now proceed with at least 5 rounds of rate increases. Be prepared.
English
90
83
1K
69.3K
allegedly!
allegedly!@januarycomputer·
@0xSero this is literally how i speak lmao
English
1
0
0
26
0xSero
0xSero@0xSero·
I think I’m going to start a series How to spot slop 1. Curious 2. Question mark 3. Balance 4. Affirming language LLMs mostly try and get to the next turn, so they will overwhelmingly ask questions like this When prompted to craft a response to x amount of visible accounts
cCross@TheCesarCross

@0xSero Tradeoffs are real, curious how you balance cost and latency?

English
8
1
29
2.7K
Lotto
Lotto@LottoLabs·
I need my 3090 to run training test, but then I can’t run my Hermes agent 27b model, gonna need another card or a cheap api
English
9
1
31
4.1K
0xSero
0xSero@0xSero·
Using my own subs for Droid is such a blessing, I still have to pay 20$ for Factory but it's well worth it. I'm going to mission doing my taxes this week, stay tuned.
0xSero tweet media
English
21
1
90
7.1K
Sudo su
Sudo su@sudoingX·
been playing with hermes agent paired with qwen 3.5 dense 27B on my single 3090 since last night. there is something about this harness that caught me and i think i know what it is. i've now run five qwen configs on consumer hardware: 35B MoE (3B active) -- 112 tok/s flat across 262K context, 1x 3090 27B dense -- 35 tok/s, zero degradation across the same range, 1x 3090 qwopus 27B (opus distilled) -- 35.7 tok/s, same architecture, different brain 80B coder -- 46 tok/s on 2x 3090s, oneshotted a 564 line particle sim 80B coder -- 1.3 tok/s on 1x 3090, bleeding through RAM because it didn't fit but it still ran with same benchmarks. same prompts. same quant where possible. every config is documented. i know these models. and hermes agent is the first harness that feels like it respects that work. tool calls show inline with execution time. nvidia-smi 0.2s. write_file 0.7s. you see exactly what the agent is doing and how long each step takes. no mystery. no black box. no tool call failures so far and i've been pushing it. most agent frameworks feel like you're watching a spinner and hoping. hermes shows the work. that transparency changes how you trust the output. once you use it you see the UX decisions are not accidental. @Teknium and the nous team built this like engineers who actually use their own tools. 80 skills. 29 tools. persistent memory. context compression. runs clean on a single consumer GPU.
Sudo su@sudoingX

okay the fuss around hermes agent is not just air. this thing has substance. installed it on a single RTX 3090 running Qwen 3.5 27B base (Q4_K_M, 262K context, 29-35 tok/s). fully local. my machine my data. first thing i did was tell it to discover itself. find its own model weights, check its own GPU, read its own server flags, and write its own identity document. it did all of it autonomously. nvidia-smi, process grep, file writes. clean execution. the TUI is genuinely premium. dark theme, ASCII art, color coded tool calls with execution times, real time streaming. you actually enjoy watching it work. 29 tools. 80 skills (that's what it reports on boot). file ops, terminal, browser automation, code execution, cron scheduling, subagent delegation. and it has persistent memory across sessions. setup took 5 minutes. one curl install, setup wizard, point to localhost:8080/v1, done. dropping qwopus for this test btw. distilled models compress reasoning and lose precision on real coding tasks. base model only from here. more experiments coming. octopus invaders (the same game that broke qwopus) will be built using hermes agent next. comparing flow and results against claude code on the same model. if you want to run local AI agents on real hardware this one deserves a serious look.

English
43
44
680
98.2K

Keşfet

@G333M4N @zerohedge @felixrieseberg @ffodaehruoytuc @zeynepmyenisey @alienated_cola @sainimatic @02020hz