The easiest way to find out which models you can run on your computer:
Just run:
- ๐ป๐ฝ๐บ ๐ถ -๐ด @๐บ๐ฎ๐ด๐ป๐ถ๐๐๐ฑ๐ฒ๐ฑ๐ฒ๐/๐ฐ๐น๐ถ
- ๐บ๐ฎ๐ด๐ป๐ถ๐๐๐ฑ๐ฒ ๐๐ฒ๐๐๐ฝ
It profiles your machine and ranks the models across:
- Speed
- Accuracy
- Intelligence
- Memory required
Finally, you can choose your favorite harness (Pi, OpenCode, Claude Code, Codex, etc.) to run with it.
Get started here: https://t.co/gpwU34NFlw
(don't forget to star ๐)
I also wrote a detailed article on running your favorite harnesses with local models. The article is quoted below.
The easiest way to find out which models you can run on your computer:
Just run:
- ๐ป๐ฝ๐บ ๐ถ -๐ด @๐บ๐ฎ๐ด๐ป๐ถ๐๐๐ฑ๐ฒ๐ฑ๐ฒ๐/๐ฐ๐น๐ถ
- ๐บ๐ฎ๐ด๐ป๐ถ๐๐๐ฑ๐ฒ ๐๐ฒ๐๐๐ฝ
It profiles your machine and ranks the models across:
- Speed
- Accuracy
- Intelligence
- Memory required
Finally, you can choose your favorite harness (Pi, OpenCode, Claude Code, Codex, etc.) to run with it.
Get started here: https://t.co/gpwU34NFlw
(don't forget to star ๐)
I also wrote a detailed article on running your favorite harnesses with local models. The article is quoted below.
What if GPT Astra learns to boxing ?
Human and robots
Becoming cyborg
Knocking out in one punch
Or getting short circuited by human punch or Virus Punch๐ญ๐ญ
Can't believe this is free.
โ https://t.co/G9QFJqwHBq
2,000+ hand-drawn illustrations you can use for your projects.
Download them as:
โข PNG
โข SVG
Mix and match them to create full-blown scenes.
Absolutely insane resource.
I wonโt think anything crazy happened than thisโฆ
- Guys after 1 month of a dead declared laptop, suddenly booted after smashing in rage ๐ค
- got my NPM access to publish my packages
Introducing TimesFM-3, a state-of-the-art time series foundation model that enables accurate multivariate time series forecasting in a single forward pass, significantly outperforming other forecasting models across major benchmarks.
More on the blog โhttps://t.co/uSlnIdUJ4Q
Hey @X , connect me with 100 curious builders, researchers, and AI practitioners to whom I can hand the infinity gauntlet of future of AI.
Iโm launching a private AI workstation for running and benchmarking local models on your Mac.
Built for Hardcore Engineers, Researchers, ML, RL, DL, kernel, open-source AI enthusiasts grinding day and night for upliftment of humanity.
Early access form coming soon.
Stay tuned.
@H23ko5723323@OpenAI@FarzaTV@heyclicky 100%.
Models commoditize โ coding agents commoditize โ execution becomes the moat.
Reliable computer control + permissions + memory + an ecosystem of app-specific skills could basically become the OS layer for agents.
Thatโs the interesting battlefield now.
What the hell @OpenAI ๐ญ
Codex can control my Mac now.
@FarzaTV@heyclicky, the competition is no longer just โwho has the better coding agent.โ
Itโs becoming:
Who owns the computer-use layer.
One possible moat:
Build a marketplace for AI Skills.
Creators should be able to teach an agent exactly how to use Figma, Blender, Excel, Premiere, Shopify, literally any app, and get paid per install or execution.
Vision gives AI eyes.
Community-built skills give it muscle memory.
The real opportunity may not even be the skill itself.
Itโs the economy around skills:
distribution, permissions, benchmarks, trust, versioning and monetization.
Eventually, experts could just record how they work โ AI turns it into a reusable skill โ thousands of agents can use it.
Basically:
monetize how you work.
Ship it before @OpenAI owns that too ๐ ( Or I can help you with the plan and execution as I am a builder too )