I build AI agents for a living and write about it. Writing at https://xitter.twitterapi.xyz/t.co/ROXTvUfk75 · Writing @latentspacepod · Agentic Ops @ AnswerRocket
United States
Joined October 2013
- Tweets38.2K
- Following4.8K
- Followers33.1K
- Likes73.6K
Pinned Tweet
I shouldn't do this, but I'm gonna let you guys in on a lil secret:
Claude Fable 5.2 + '/be-ambitious' (link 👇)
Change your life today.
Gemini 4 Hydrogen is the Gemini 4 model family I'm most excited about. The Goog's small models routinely exceed expectations.
5
36
2,170
Either GPT-6.2 Astra / Sol absolutely CRUSH tomorrow, or I cancel my ChatGPT sub.
Sorry I'm not sorry.
Wtf are they feeding this thing???
Claude Fable 5.2 (Max) is now #1 on Agent Arena, leading the Pareto frontier in real-world agentic performance!
Top models on Pareto Frontiers:
- #1 Claude Fable 5.2 (Max): +30.4% | $7.3/task
- #2 Claude Opus 5.5 (Max): +19.5% | $5.5/task
- #3 Gemini 4 Carbon (High): +17.0% | $1.5/task
- #7 GPT-6.1-Sol (Max): +12.5% | $1.1/task
- #8 GPT-6.1 Terra (Max): +10.8% | $0.9/task
Fable 5.2 delivers over 1.5x the net improvement of the #2 model, Opus 5.5, at a 32% higher median cost per task. Compared to Fable 5.1, it also delivers roughly 2x the net improvement at a 6% lower cost.
Congrats again to the @AnthropicAI team on this impressive release!
6
108
14,905
My workday consists of one repeated decision: delegate this task to an agent or not?
Can I give the agent enough context? Is this something I enjoy? Do I want to give it up?
It's exhausting.
Claude giveth 🙏
We've given all Pro, Max, Ultra, and Team users one free full reset, across all models and limits. Use it anytime before December 10 at 2:00PM PT.
We recently fixed a few small issues that used more limits than intended, and the Fable resets didn't do much for those of you who'd already hit your weekly limits for all models. While these issues didn't affect everyone, we want to make up for the trouble.
You'll find it in Settings > Usage, or at clau.de/reset.
Really fun read. No one in AI is better at predicting the future. 10/10 recommend.
I agree with everything besides the implicit negative valence and neuroticism borne from the conclusions.
Unfortunately, it seems like the intelligence explosion will be happening in 2027 after all. Most of this post analyzes the new METR report (though other sources of evidence are also important). If they’re right, that implies an AI-driven acceleration that is 5 months ahead of the AI 2027 scenario. I am scared, sad, and angry. blog.aifutures.org/p/q35-tim…
An intelligence explosion in *2027* is highly likely, according to the new AI Futures timeline. It has become undeniable.
It's all really happening. Everything the best science fiction writers imagine. Unfathomable advances in medicine, biology, energy, mathematics and more.
Technological advances that we can't conceive of yet, too.
It's a once-in-a-civilization level opportunity. The only way to screw it up is if we resist.
Fable 5.2 prompt audit skill to get your project READY.
Here's what it does, all the tips from Anthropic's official guide on prompting Fable 5.2:
- Finds standing instructions to delegate to subagents
- Finds standing token budgets
- Finds harness instructions that edit earlier turns (Fable 5.2 uses "preserved thinking")
- Finds Opus 5.5 prompt that instructs the model to "treat the answer as done"
- Finds instructions to think less
- Finds rigid style rules
- Finds rules, notes and memories that make Fable 5.2 overly cautious
Install '/fable-5-2-prompt-audit' at the repo below.
“Mom, how did everyone get so poor?”
“Everyone was heavily invested in NVIDIA and the CEO decided to get honeypotted by a Chinese spy.”
4
17
1,687
ANT to IPO at $70 - $78 per share.
Shares available to trade estimated Thursday / Friday next week.
Anthropic has launched the roadshow for its proposed initial public offering: anthropic.com/news/roadshow-…
Legal disclaimer: A registration statement relating to these securities has been filed with the SEC but has not yet become effective. These securities may not be sold, nor may offers to buy be accepted, prior to the time the registration statement becomes effective. This post does not constitute an offer to sell or the solicitation of an offer to buy these securities, nor shall there be any sale of these securities in any state or jurisdiction in which such offer, solicitation or sale would be unlawful prior to registration or qualification under the securities laws of any such state or jurisdiction.
3
17
1,518
Opus 5.6 aka ‘Model 4’ from the METR RSI report. My guess.
Opus 5.5 was the most capable Claude model for a while, outperforming Fable 5.1. It’s totally plausible.
Finished training last month.
