Introducing Codex++
The power user codex-cli fork:
- Multi Account Support
- Auto-Start Weekly Limits
- Auto-switch to new account when empty
- Guardian-review with --yolo based on hooks
npm i @jjliebig/codex-plus-plus
I used ~20b 5.6-sol tokens with it, it's very fun.
Replying to @grhmc @ImLunaHey
I love the "later integration" approach here, very European
Replying to @dominikkoch
I got my own ci pod from OVH, 170$ a month and already 10x savings vs GitHub. And I'm penny crunching for a self funded Start-up.
Can be scaled pretty easily, only thing I recently had to fix was artifact storage and even that is pretty easy.
Honestly a solid option
Replying to @petergostev
Isn't that with model training included, so inference is way way way cheaper even?
Replying to @skalskip92
This checks out, Terra has been unbelievably good one some very clear tasks I threw at it like PR review
But the PRs it sends itself are pure trash compared to Opus or Sol.
Odd little model, very fun to use though
It's gonna be hilarious when everyone looks back 6 months from now and remembers comparing cost per token and how stupid that was.
Or comparing subscription API equivalent costs and how stupid that was.
The only thing that matters is intelligence per $.
What the fuck
BREAKING NEWS🚨: NVIDIA just submitted its first Rubin NVL72 Kimi K3 submission, and it can reach over 850 tok/s/user, even at 2.8 Trillion parameters, with over 3x higher throughput on @vllm_project!
Great work by the NVIDIA team, and excited for Rubin Software 1.0 GA coming soon!
I mean note the error bars and that's likely just Terra coming out of the gate swinging but shesh
Actually very impressive first showing for reviews.
well played @reach_vb & gang
arena.liebig.gg - give it a few more days and it should have Terra more grooved in
This is heresy, the emperor will not be amused
good read for what might come next:
emanueleviola.wordpress.com/…
Replying to @kimmonismus
There is no chance
Not after they killed Astra 6.1
That's cope and y'all know it (sorry)
The smart strategic play is to let anthropic win this round, huge IPO and then fight back after so ai market looks juicy still
very curious how well it performs for reviews, Sol is the clear king, Opus is legit just... insane (if you loop reviews until green) but $$$
so this one might be the reviewer of choice for plus plan enjoyers
Replying to @theo
pretty sure Luna is so much worse than Mimo 2.6, it's actually not bad
5.6 Luna was a treat, 6 Luna is.. idk, pretty bad honestly
Replying to @hthieblot
Stronghold.
Not crusader, fuck that gold spam shit.
And Wiggles. That one was insane
BRUH
Replying to @aliceisplaying
I think “attacker must be on the tailnet” is an important detail here worth including
Replying to @Angaisb_
different one, the instant-compaction was something else, this one is (iirc) different and new











