tweetindex
ID

Chubby♨️ ✓

@kimmonismus · Germany · joined 23 Sep 2022

Tech Analyst & Content Creator. Editor-in-Chief @getsuperintel - Community of 300k+ in total: 🔗 https://t.co/jHMmImmI5I //📧 kim@getsuperintel.com

145 834Followers
3 374Following
44 662Posts total
4.7MViews on collected posts

Postingan terbaru

I mean, it has to be this way, right? The Opus 5.5 doesn't just feel different, it's incomparable to the Opus 5. The 5.5 is smaller, cheaper, more intelligent, and has some amazing taste. Distilled from an internal Mythos-2. Makes the most sense. 7.2K views · 114 likes · 0 reposts · 9 replies Open on X →
Meta Connect started. I’ve already had a chance to see a few things, but more on that tomorrow when things really kick off. I’m really excited about Mark Zuckerberg’s keynote. Connect features some exciting events, like this AI-generated scenario by an artist. Oh, by the way
2:34
12.7K views · 210 likes · 6 reposts · 9 replies Open on X →
The more I use, the more I fall in love with opus 5.5. its so good and so much faster and so much less verbose. It’s everything I could have asked for. It’s as if my beloved Opus 4.6 came back but better and rebranded as 5.5 63.1K views · 2.4K likes · 59 reposts · 78 replies Open on X →
A quick recap of what we saw today, and an answer to the question of whether there was a winner: both OpenAI and Anthropic gave us plenty to be excited about. Let’s start with Anthropic. They addressed feedback about how Opus communicates, and they’re giving subscribers a reset
33.6K views · 547 likes · 30 reposts · 37 replies Open on X →
Opus 5.5 was obviously trained by a much bigger teacher model. Probably Model-2 Mythos. Opus 5.5 is the first model trained from RSI and also distilled from the internal Ant teacher model. This is how Opus 5.5 is both smaller and cheaper. It is an artifact of teacher model 299.1K views · 3.5K likes · 131 reposts · 61 replies Open on X →
@kimmonismus @synthwavedd surely this means that Fable 5.5 is coming soon? 1.4K views · 8 likes · 0 reposts · 1 replies Open on X →
@synthwavedd Absolutely insane evals so far. 6.1K views · 56 likes · 1 reposts · 2 replies Open on X →
And here are the benchmarks! Opus 5.5 excels Fable 5.1 in literally every single eval and even GPT-6 Astra in almost every single one. Holy moly, this is going to be interesting! h/t @synthwavedd https://t.co/4OD6mZXVFC
86.2K views · 1.6K likes · 79 reposts · 59 replies Open on X →
Opus 5.5 already in Claude Code! Release imminent! https://t.co/tXYHbp9AEQ
99.1K views · 512 likes · 8 reposts · 35 replies Open on X →
@kimmonismus @kristy_loke Thanks for reposting our story 🙂 1.3K views · 8 likes · 0 reposts · 2 replies Open on X →
Source: https://t.co/kgywzGUwmc 7.2K views · 24 likes · 0 reposts · 0 replies Open on X →
A secretive Chinese AI startup backed by Tencent and valued at $1.42 billion is reportedly preparing to release its first open-weight LLM as early as this month. Via The Information: Tsinghua professor Jifeng Dai founded the startup in February. Its approach: take an existing
54.6K views · 1.1K likes · 71 reposts · 40 replies Open on X →
@kimmonismus The next GPT model will actually release faster than GPT-5.6 Sol > GPT-6 Astra 1.2K views · 28 likes · 0 reposts · 1 replies Open on X →
@kimmonismus I don't care. If they just focus on making astra faster and cheaper, that would be fine 7.1K views · 24 likes · 0 reposts · 2 replies Open on X →
@kimmonismus The previous slowdown was just 2 weeks. We shall see if that's the pattern or it will continue to increase 7.3K views · 56 likes · 0 reposts · 1 replies Open on X →
I’m seriously worried that GPT-6 Astra may have been the last release to follow its predecessor so quickly. GPT-5.6 Sol launched on July 9. GPT-6 Astra began rolling out on September 3. Exactly eight weeks apart. With the push to slow down AI development, I worry this may have 160.8K views · 2.4K likes · 47 reposts · 261 replies Open on X →
One of the most reliable posts. I wonder when Anthropic will finally address this problem. They manage to solve the most difficult scientific problems with their models, but they can't seem to make their models more efficient or less verbose. Absurd. https://t.co/LJfYPh3BUb
87.3K views · 1.7K likes · 58 reposts · 95 replies Open on X →
Another Codex reset incoming. I wasn't even prepared for it because I wasn't expecting it. If Anthropic is resetting now too, then OpenAI will probably have to reset twice as often. I like it. :) 62.1K views · 953 likes · 13 reposts · 39 replies Open on X →
Interesting release: OpenAI says ChatGPT Work can now learn your writing style from connected emails, messages and files. "Your phrases, sign-offs, even capitalization quirks, carried into future drafts." This gives non-human work processes, now carried out by agents, a https:/
42.1K views · 544 likes · 16 reposts · 34 replies Open on X →
The Website traffic share provides a fairly accurate indication of when OpenAI's turnaround occurred. Around two months ago, roughly the same time as Codex/ChatGPT Work experienced its exponential growth, OpenAI's traffic share relative to its competitors increased for the first
32.7K views · 307 likes · 20 reposts · 26 replies Open on X →
@kimmonismus more context isn’t necessarily better context. the challenge is making the right connections without letting irrelevant history influence the result. 540 views · 5 likes · 0 reposts · 1 replies Open on X →
@kimmonismus We made AI insanely good at doing things. Still waiting for the part where it knows which things are actually worth doing. 286 views · 3 likes · 0 reposts · 0 replies Open on X →
@kimmonismus the best AI won’t need constant steering. it’ll know enough about you to move first 275 views · 2 likes · 0 reposts · 0 replies Open on X →
An AI can complete a task perfectly and still spend your time on the wrong thing. Imagine an assistant preparing notes for tomorrow’s meeting. A message arrives that changes the deadline, and suddenly another project needs your attention first. The notes may be excellent. The ht
160.8K views · 155 likes · 9 reposts · 32 replies Open on X →
@kimmonismus lol https://t.co/ZUAhmluyRI
3.1K views · 55 likes · 0 reposts · 0 replies Open on X →
@kimmonismus i have not yelled at it in over 3 hours, shocking tbh 11.9K views · 133 likes · 0 reposts · 7 replies Open on X →
@kimmonismus no. it is great but a usage killer am on 20x pro and it consumes like 4-5%/hour feels like right now, on medium but otherwise a really great model 4.2K views · 55 likes · 2 reposts · 9 replies Open on X →
Crazy, I haven’t seen a single negative post about Astra. Not here, not on Reddit. Everyone’s praising its speed, token efficiency, intelligence, and ability to get things done. This might genuinely be OpenAI’s best release ever. 262.2K views · 6.6K likes · 222 reposts · 378 replies Open on X →
@kimmonismus But where is it? And how come benchmarks are coming out without an OpenAI announcement..? What the hell is going on? 😂 9.2K views · 100 likes · 1 reposts · 0 replies Open on X →
https://t.co/AHH4LH0Q3a 43.8K views · 98 likes · 4 reposts · 0 replies Open on X →

Dibandingkan akun berukuran sama

55 postingan dari 90 hari terakhir, dibandingkan dengan rentang 100K–1M pengikut. tampil jauh melampaui audiensnya sendiri, dan audiens itu merespons.

Median tayangan7 274akun ini5 983median untuk 100K–1M
Jangkauan, %4.99%akun ini1.49%median untuk 100K–1M
Interaksi, %1.17%akun ini0.93%median untuk 100K–1M
MetrikAkun iniMedian untuk 100K–1MRasio
Median tayangan per postingan7 2745 9831.22×
Jangkauan (tayangan ÷ pengikut)4.99%1.49%3.35×
Tingkat interaksi1.17%0.93%1.26×

Akun lain pada rentang ini →   Bandingkan dengan akun lain →   Bagaimana tolok ukur ini disusun →

Growth & engagement

How the posts we collected actually performed: views and reaction rate post by post, what the audience did with them, and where the follower count goes.

Views per post

7.1K13 Sep
1.2K
54.6K18 Sep
7.2K
1.3K
99.1K22 Sep
86.2K
6.1K
1.4K
299.1K
33.6K
63.1K
12.7K23 Sep
7.2K

Last 14 collected posts, oldest on the left. The scale is logarithmic: one post can outrun the rest a hundred times over.

Engagement rate per post

0.37%13 Sep
2.42%
2.20%18 Sep
0.33%
0.78%
0.56%22 Sep
2.02%
0.96%
0.63%
1.24%
1.83%
4.05%
1.78%23 Sep
1.70%

Reactions — likes, reposts, replies and quotes — divided by views. Median for 100K–1M accounts is 0.93%.

What the audience does

Likes84.6%23 324 in total
Reposts2.8%777 in total
Replies4.4%1 219 in total
Quotes0.5%131 in total
Bookmarks7.7%2 127 in total

Share of every reaction we collected for this account. Replies mean argument, reposts mean endorsement, bookmarks mean the post was worth keeping.

The follower curve appears once this account has two daily snapshots — we take one a day, and this one is on its first.

Akun serupa