Most people using AI agents right now are managing chaos.
- Too many tabs.
- Too many shells.
- Too many half-connected tools.
- No shared context.
- No guardrails.
- No operating model.
We built CoWork OS for that exact problem.
๐งต๐
@shannholmberg Full respect to Hermes team to also go into this direction! In @Cowork_os we are building the agent gui-first from day one and its much better and much more developed!
@billgnofficial building open source alternative (named @Cowork_os) to codex, Claude code and cursor where you can use any LLM provider you want, even local llms and the harness and features are as good if not better!
@Jason that's why I'm building open source alternative (named @Cowork_os) to codex, Claude code and cursor where you can use any LLM provider you want, even local llms and the harness and features are as good if not better!
@NousResearch great to see that Hermes is also have GUI now! congrats. Also full respect to Hermes agent and Nous Research team, if you're looking for a GUI-first agent you can check CoWork OS too!
codex's 2x usage limit promo ends on may 31st
that's in 4 days
if your usage already feels like it evaporates too fast, get ready for june - it's going to be rough
openai is willing to give eligible enterprise customers 2 months of free codex usage
but pro users are about to lose the extra headroom that made codex feel like a workhorse
so i'm hereby creating a petition
dear @sama please make the 2x codex limits permanent for pro subscribers - you know it's the right thing to do
sign my petition by liking this post and sending it to your mom
@1752vc not raising but still wanna share the super (everything) app I'm building, much better than openclaw and similar offerings and has more features than codex where you can use it with any llm provider including chatgpt and claude subscriptions
@AndrewCurran_ They had the biggest chance to be the stars of this age, but theyโve been a real disappointment so far. They can still turn it around though; itโs not too late.
1/ CoWork OS v0.5.47 is out.
This release focuses on something that becomes painful as agents scale: long-running session stability.
Less renderer churn, lower memory pressure, safer permissions, more resilient recall, and smoother multi-agent execution.
https://t.co/GDgN7H8L0b
GPT 5.6 completely destroys Gemini 3.5 Pro btw
It's not even close, sorry if anyone expected something SOTA from Google
I think this will be true for the next Anthropic release as well, obviously.
THX @synthwavedd for showing me the way
@emollick You cant use claude models in codex and gpt models on claude code + cowork! You can do that on @Cowork_os though which also supports all other llms like kimi and deepseek etc via either api or subscriptions and its opensource ๐