Then you try to pin discount pricing when everyone has some discount going on, and knows they will release some new model before it ends anyway.

And if you’d have used Muse Max at all you’d know it’s not been really in the same league as the rest. Then again Astra and Sol are both not really competing with Opus now with 5.5 either.

It’s a solid release on paper and until anyone uses it I find anyone trying to proclaim a loser completely unreliable.

If they’re smart they pull a DeepSeek and teach flash to beat this entirely within a month or two, I wouldn’t bet against it.

It reset just a few hours ago and I've been running it, couldn't be more than 15 sessions none more than an hour long:

---

Session usage: no model calls yet in this session.

Weekly limit: 46%

Next reset: September 27, 23:20

---

I actually have to believe my account is messed up tbh, it's so bad. For reference I've ran 12 fable and some ~40 Opus sessions since reset yesterday on a CC account, at least 5x more usage by my estimate:

  Current week (all models)

                                       28% used

  Resets Sep 28 at 5am (Pacific/Honolulu)

Ok looking at it more, Grok and Grok Build just really suck. They are about 10x less token efficient, often using 200+ tool calls in a row for what are not even big tasks where Opus would use 5-10. Their cache hit rate is worse, and two sessions got into basically unnecessary loops costing a solid quarter of the entire week. And this was on smaller tasks as I tend to use it for easier things.

Me 78 days ago - https://news.ycombinator.com/item?id=48693623

And see the commenter agreed with me.

Lol @ sus though, I mean I am curious how this can be because I do see people saying Codex is more generous and wonder how it can be. I have too much usage for too long to have any doubts, but for all I know OpenAI black boxed me or Anthropic put me in some nice bucket, I wouldn't be surprised if they do that. I did see some mention that they can limit your tokens if they suspect you of things, though I forget the source of that.

I've verified this with the heaviest users I know, I've run the numbers myself over and over.

I have too many max accounts on each to not know this. My Claude accounts typically are doing 2x the number of sessions, and every single week the Codex accounts run out faster even with all these resets. I can easily burn a full weeks usage in a half day, it's closer to 1.5 with CC.

There was only a brief window of time that the opposite was true.

And as for outdated dependencies, breaking changes, and unfriendly companies, the reason I ever got into RN was after trying to build an iOS app using SwiftUI (and then bailing to UIKit to no avail). Talk about a borked ecosystem.

The frameworks, developer tools, languages, and iteration speed of RN hands down beat any one platform, and you have one model for your data and state which is an unequivocal win. I agree it has plenty of warts, but the arc of progress is that they have been going down steadily.

If you're a 3k person developer team where 200ms can cost millions a year, then by all means go purely native. Though I bet if they'd just gone to the new architecture and spent 1/5 of the cost in agents optimizing things and building a few native bridges they'd have nicer apps and be shipping faster.

This is well established, not at all a question.

> they might reduce some inflammation markers slightly more than weigh loss alone

There's no "slightly," It's going to be a news story for years to come. It's a broad spectrum anti-inflammatory drug of a breadth and efficacy never seen before.

Reading the DS reasoning is wild, it's constantly going in circles. The most minor lack of clarity in your prompt and it will spend ages going back and forth on what you meant. It reasons 5x longer than the preview which makes it really slow now as well. We did a lot of work to nudge it to be decisive and improve our evaluation setup to there's more clarity, and it helped but only marginally.

Ours is a full-stack app one shot test so it includes backend, frontend, design, and QA/testing. It's graded by Opus xhigh and Sol xhigh and the grades are averaged.

DS4 preview would finish in 20 minutes flat on high reasoning and grades 6/10. Luna high gets 9/10 in about 30 minutes. DS4-final is crazy - at high thinking it's taking over an hour and getting ~8 but only had one successful run as I got tired of waiting so long after many early abort/retries trying to debug why thinking was so long. The lowest thinking still takes over 45 minutes, and with thinking off it actually is finally closer to preview in time but actually get's a much more varying result anywhere from incomplete to 6 it seems.

Costs per run DS4 is best but not actually by a lot as it's spending 10x the tokens with all the reasoning and mistakes. It's a very brute force model and I really preferred preview in many ways for how predictably fast it was.

Side note, Spark 1.2 is a nice model for this test, best in frontend design and fastest to get results together, though not nearly as efficient as Luna. Grok scores similarly to Spark but at like $50/run vs the contributor Spark costing $1.50.

Edit: was curious to see and seems DeepSWE agrees at least: https://www.together.ai/blog/deepseek-v4-flash-0731-vs-gpt-5...

Edit 2: btw it tests a team of agents working together in a special harness and stack. So 20 minutes is for 8 agents essentially. That said everything was built around DS as it was the cheapest to iterate against so even with that advantage the new one struggles.

Spark is actually the interesting one imo. It's significantly better, also significantly faster. If you are ok with letting Meta soak up your data (which DS does too) it's also the same price.