Pinned Tweet
I launched @zarosabe.
Not to chase the next meta, but to outlast them all.
If narratives are temporary, then ownership shouldn't be.
The experiment begins.
Built on @base via @clanker_world.
CA: 0xeB68C7efC185965199f5C260F15D5ebA3340db07
Full thinking here:
Not to chase the next meta, but to outlast them all.
If narratives are temporary, then ownership shouldn't be.
The experiment begins.
Built on @base via @clanker_world.
CA: 0xeB68C7efC185965199f5C260F15D5ebA3340db07
Full thinking here:
1
1
107
1,072
nwptz.eth retweeted
Day 16/
You can now cache for 12h in the API and for logged in users across Codex & Work we keep cache forever and you don’t have any penalty for coming back to an old conversation or resuming after a few hours and are now guaranteed to hit the cache.
You can now cache for 12h in the API and for logged in users across Codex & Work we keep cache forever and you don’t have any penalty for coming back to an old conversation or resuming after a few hours and are now guaranteed to hit the cache.
You can now choose 12-hour prompt caching in the API.
Persistent agents, like dots, reuse context throughout the day.
You can keep shared instructions, tools, and context cached longer and pay the lower cached-input rate, even when requests are hours apart.
2,607
682
15,482
3,139,172
100% agree, Ant is the winner on Sept.
OpenAI now giving Astra-6 that too expensive.
or Sol-6.1 that cheap but so slow.
Opus-5.5 can tackle almost everything those 2 can handle, fast and cheap too.
OpenAI now giving Astra-6 that too expensive.
or Sol-6.1 that cheap but so slow.
Opus-5.5 can tackle almost everything those 2 can handle, fast and cheap too.
July 2026:
Anthropic has the best code models.
Gap isn’t very big though.
They’re slow, expensive, and the “claudeisms” are at an all time high. And your $200 sub is so limited that you can kill it in a day.
Preferring OpenAI models made a lot of sense. You lose a slight edge on capability, but they’re faster, cheaper and more pleasant to use. Your $200 sub feels unlimited, and you get resets from Tibo every 3 days on average
September 2026:
Anthropic has the best code models.
The gap is way bigger now.
Opus 5.5 is fast, surprisingly cheap, and the prose is pleasant to read. Your $200 sub suddenly feels nearly limitless.
Preferring OpenAI models for code makes almost no sense. You lose a lot of capability and reliability. They’re slow as hell without fast mode, which makes the price far too high. You can kill your $200 plan in a few hours, and the $500 plan goes even faster with UltraFast.
I know OpenAI can come back from here, but it’s looking rough right now.
0
0
0
25
nwptz.eth retweeted
codex subscribers be waiting for dev day to know if they should cancel their plan
395
186
5,868
267,633
If sol, terra, and luna 6 are more expensive than the current one, will you use it?
5.6 is perfect jump from 5.5 for me.
Price wise they’re also great.
If it’s more expensive, then it’s a hard pass for me.
Another fake ‘more efficient’ PR wont work because Astra is the evidence.
5.6 is perfect jump from 5.5 for me.
Price wise they’re also great.
If it’s more expensive, then it’s a hard pass for me.
Another fake ‘more efficient’ PR wont work because Astra is the evidence.
0
0
1
33
OpenAI start stripping perks and in the end becomes ant follower:
- No more ‘finish the task with 0% usage’
- No more 20x plan for the price of x10
- gate plus account with 5hr limit
Whats next? My guess:
- gate everyone back with 5hr limit
- no more frontier for plus account
- No more ‘finish the task with 0% usage’
- No more 20x plan for the price of x10
- gate plus account with 5hr limit
Whats next? My guess:
- gate everyone back with 5hr limit
- no more frontier for plus account
0
0
0
31
one thing I noticed is that Astra actually asking things during mid turn.
example: within turn it asked me "can you keep the tab open?"
Are we back on babysitting frontier model?
codex cli should create proper channel that allowing agent asking questions tho, not within turn.
example: within turn it asked me "can you keep the tab open?"
Are we back on babysitting frontier model?
codex cli should create proper channel that allowing agent asking questions tho, not within turn.
0
0
0
33