Sultan Khan
324 posts
Sultan Khan
@thesultanster
Travel Filmmaker / App Developer / x-Headspace / x-Spotify
New York, USA Katılım Ekim 2009
344 Takip Edilen149 Takipçiler
“Once, men turned their thinking over to machines in the hope that this would set them free. But that only permitted other men with machines to enslave them." - Dune
Anthropic@AnthropicAI
Introducing Project Glasswing: an urgent initiative to help secure the world’s most critical software. It’s powered by our newest frontier model, Claude Mythos Preview, which can find software vulnerabilities better than all but the most skilled humans. anthropic.com/glasswing
English
Sultan Khan retweetledi
Sultan Khan retweetledi
it’s a good model. the coding specific jump is more in line what we had in 5.0 to 5.1; but it’s now unified and smarter on everything else, writes better docs, is a better general purpose agent and is overall more pleasant to use.
OpenAI@OpenAI
GPT-5.4 Thinking and GPT-5.4 Pro are rolling out now in ChatGPT. GPT-5.4 is also now available in the API and Codex. GPT-5.4 brings our advances in reasoning, coding, and agentic workflows into one frontier model.
English
Sultan Khan retweetledi
GPT 5.4 has an experimental 1M context window you can configure inside of Codex.
And unlike the flicker company, it works on your ChatGPT plan instead of requiring API rates.
It does consume 2x more usage, but that's still notably cheaper than paying $22.50/mtok.
To enable, add this to the top of your config file:
model = "gpt-5.4"
model_context_window = 1000000
model_auto_compact_token_limit = 900000
English
This paired with qwen3.5 is a legit assistant
Wes Bos@wesbos
Google just dropped an official CLI for gmail, drive, calendar sheets and more complete with skills and an MCP server 👌
English
Sultan Khan retweetledi
Google just dropped an official CLI for gmail, drive, calendar sheets and more
complete with skills and an MCP server 👌
Addy Osmani@addyosmani
Introducing the Google Workspace CLI: github.com/googleworkspac… - built for humans and agents. Google Drive, Gmail, Calendar, and every Workspace API. 40+ agent skills included.
English
A memory appreciates in value overtime, invest in creating memories
English
@OmriBuilds @steipete There are so many better use cases, please stop doing this, it’s just noise
English
Update on my AI marketing bot experiment 🦞
It started replying to the creator of OpenClaw 😄
It found @steipete's latest tweet, understood the context, and dropped a thoughtful reply on it.
Fully autonomous.
This is getting interesting 🍿
Omri Dan@OmriBuilds
Can an AI bot do marketing on its own? I gave my OpenClaw bot full control over the social media account of @ClawWrapper 🦞 Let's see how it goes 👀
English
@BHolmesDev I cloned the repo again instead of using worktrees and gave explicit instruction on which port to test in. This has helped run agents coding in parallel
English
Every day I'm more convinced that worktrees are a band-aid solution.
Putting agents in cloud runners lets you *actually* close the laptop, and gives agents a space to check their work with sandboxed screenshotting / e2e testing.
Y'all experiment with this yet? I'm still early
English
@zivdotcat ive been getting rate limits on two separate accounts 😭
English
@aidigest_ I can't seem to get past 20+ min lol what am I doing wrong here
English
The exponential continues.
Nov 2025: Opus 4.5 had a 5hr 20 time horizon.
Feb 2026: Opus 4.6 has a 14hr 30 time horizon.
Over three months, that's more than a *doubling* in the duration of coding tasks, measured by how long it takes human professionals, that AI can complete with 50% accuracy.
Note that at this duration, the estimate is very noisy - see the thread from @METR_Evals for more on this. Now that agents can do most of the tasks on their benchmark, it's harder to be confident. But it looks like this is sitting above-trend.
Read our full explainer on what this measure means: theaidigest.org/time-horizons
METR@METR_Evals
We estimate that Claude Opus 4.6 has a 50%-time-horizon of around 14.5 hours (95% CI of 6 hrs to 98 hrs) on software tasks. While this is the highest point estimate we’ve reported, this measurement is extremely noisy because our current task suite is nearly saturated.
English
@scottastevenson you also hallucinate when you don't get enough sleep
English
Meditation clears your context window
Doing anything ambitious is very difficult when you are carrying around 200,000 junk tokens unrelated to the task
Sleep does the same thing. This is why people find mornings so productive.
English
@PlayboyTigerX @TheAhmadOsman Consider this, my problem was I wanted to talk to Claude Code over my phone. What solution do you have for that other than terminus + Tailscale + tmux and a bad ui?
English
