Addicted to creating project but never shipping them...
here you will see the ones that lived long enough to see the light...
visit my link to contact me!
To test the new codex cloud, I have given it a task that will really show how long this can run for..
its a 100% legit task, not something stupid but a true long task that would of taken me a week to complete prompting it..
so if this task is finished in some what decent results..
It will be one of the greatest features openai has released by far!
Introducing Murmur 🎉
Your FREE local dictation tool which has most of the features that Wispr Flow has but without the limit or the price tag!
opensource as well 😉
GPT-6.1 Sol is our most demanded model pretty much ever both across both the API and subscriptions.
Within ChatGPT & Codex, we were under heavy load, but have brough more capacity online and the speed should get much better in the coming hours, reaching almost twice the speed compared to what we served yesterday.
We have improved the app To support streaming model!
Which increases the speed For dictation.
And now, It's as good as wispr flow Completely local and completely free.
I developed this completely using Codex Cloud and it was amazing!
GPT 6.1 sol is good enough to do the tasks. It's a real good daily driver model.
Paired with this cloud feature. I don't really need to keep my computer on anymore..
the usage of Opus 5.5 is actually insane.. left claude code running over night, took 2 hours to complete a task and I woke up to my weekly usage being only at 3%
while astra is at 35% used.. for much less work..
Just got the $20 claude code plan and tried the motion graphic test on one of my apps to see the results..
Let me tell you.. This is A LOT better than I thought..
here is the break down. I started the prompt at %2 5 hour limit usage and finished it at 12% 5 hour usage with 2% of my week being used..
THIS IS NUTS!! a $20 plan competing with my codex $200 plan??
Also the results are not bad.. dont get me wrong, it used the wrong UI for the app.. the animation is kinda over the top as its not really applicable to my app BUT... it was done with 12% usage!!
Just got the $20 claude code plan and tried the motion graphic test on one of my apps to see the results..
Let me tell you.. This is A LOT better than I thought..
here is the break down. I started the prompt at %2 5 hour limit usage and finished it at 12% 5 hour usage with 2% of my week being used..
THIS IS NUTS!! a $20 plan competing with my codex $200 plan??
Also the results are not bad.. dont get me wrong, it used the wrong UI for the app.. the animation is kinda over the top as its not really applicable to my app BUT... it was done with 12% usage!!
I didn't believe it but this was actually one-shot.
Opus 5.5 on xhigh effort: "make a dynamic 15-second motion graphics video that shows what an incredible motion designer you are, like it's your showreel for a résumé. go all out."
I got this much usage shown, its the consistent usage but definitely my sub is reduces in capacity.
cuz before I never ran out.. now I am running out.. you can see 2 days with 0 activity due to 100% usage..
I’m growing an anonymous account from 0 followers to viral.
Step 5: get advice from my new followers ❤️
I checked he doesn’t follow me here so he won’t see this 😂
I am trying to match this guy in replies and getting a minimum of 100 replies
But I got nothing on my feed other than companies and engagement bait
Need help finding other real people
hey guys! I am trying to login into cursor but cuz my account is link to both my Google and X account..
I keep getting this error and cant log into grok bot..
when I got to my settings this shows up, other pages work fine tho..
@cursor_ai@amanrsanger@sualehasif996
so from your analytics, I noticed that you got over 100 replies
I am averaging much lower than that. so I see that is something I need to work on then..
Bro, you gotta read carefully with these releases...
That new harness is comparing the model with x-high vs max which is obviously gonna be a price difference..
its only showing that xhigh can perform as good as max which is nothing to do with harness. the max settings is a duration adjustment. it basically makes xhigh run for much longer.
someone please go and check on @thsottiaux
I think he is sick..
brother hasnt pressed the reset button in a while..
must be tired from all that resetting that he used to do..
herdr is just a tmux replacement, so you open herdr, then use whatever agent you got..
I use codex, opencode and others and they are all managed on here, you can see when they are working, pending, stopped, etc.. and you can build a plugin for a widget so you can see it on the top of your screen.
Fable 5 is showing up in games, Three.js map previews, 3D modeling, ads, image/video generation, and editing. Here’s a quick look at the examples—and how they compare with Astra.
I've used Meta's Muse Spark 1.3 for a week, and it's become my default model. It's fast, capable, and much cheaper than the top tier. Meta might finally be back in the AI race.
You seeing this @Teknium ?!?
this is why I been asking for multi connection for so long!
its AMAZING that I can control all of my devices in one interface and never have to worry about it.. I got 5 devices running hermes doing bunch of different things.
Next step - Android app PLEASE!!!
PS: All love bro!
Not sure which option you have, would love to know more but from what I can see, the same equivalent server is still the same price as the others but it's slightly cheaper..