Marcos

@starakaj this is yours?

@carlkolon Definition of Done

One of the biggest remaining problems with LLMs is that they don't know what needs to be done

@mangonkowli @_TommyMason @Ferrari yup, thinking the same, this is like google buying windsurf and "making" antigravity

@_TommyMason @Ferrari what were you thinking? Better to take the L (fail fast) and go back to the drawing board. You’re not going to sell many of these as-is.

Left, Ferrari Luce $645k Right, Nissan Leaf $35k

@gonsbi es el nuevo fiat multipla?

repito. cada dia me gusta mas mi 2012 MT

@clodoan @yitong this an artist/studio concept, right?

i’ve been thinking about this for a while since your initial posts. Heavily agree with you and I keep wondering why that is. I think it’s a deeply cultural thing. if you asked italians to redesign a ford f150 they’d never capture the essence of it. it might be nicer, but the soul of that design is american, just as ferraris are italian. People end up designing their own soul, as Johnny keeps designing apple even when he’s not supposed to

rivers of ink will be spilled about the Luce design but my biggest gripe with it is that it's clearly made by people with no rage in their heart, and no sex in their veins ferrari is about love and anger in a way that wholesome california design bois will never understand

@vikaskansalHQ @antigravity refund+damages from jan to may 2026

🚀 We’re extending the $100 AI credits promo for AI Ultra users on @antigravity until June 2nd. Keep shipping, and DM me if you hit any snags with your redemption. One per account, conditions apply!

@morganlinton hey there =) a failed 219 build update got me into 'Segmentation fault' while trying to run grok, cant run build anymore =/

Watching Grok Build think is pretty fascinating. Definitely noticing a change with the releases made this weekend, its become incredibly thoughtful. And the plans are getting insanely detailed, I mean just look at the architecture deep dive it did here.

@XorDev @benjitaylor @benjitaylor

@benjitaylor Do you need shaders?

I need this on a jacket.

@skcd42 and got locked out of build 'bash: /.grok/bin/grok: Text file busy'

@skcd42 hey there, im on a very low rate connection and it seems that the cli/install.sh has some tight timeout since i keep getting 'curl: (56) Recv failure: Connection reset by peer'

Bug fixes shipping to Grok Build 0.1.219 (release notes will be available in the TUI) - fixing usage limit bugs with prompt caching - fix layout-shifted chars in kitty terminal - replace Shift+Enter w/ Alt-Enter on VTE-based terminals (GNOME Terminal, Ptyxis, kgx, Tilix, etc.) - default expand (search_replace, write_file, run_terminal_command) calls on scrollback - mixed prose + URL paste no longer silently drops prose - set_images byte: length collision losing chips - make multi-line markdown link URLs clickable across word-wrap - clear selection when expanding truncated group - remove read_file duplicate-read check that breaks after compaction

after scamming all the pro users all of 2026 till now, you ACK that antigravity is a token burner and "reset quotas", what is going to happen after this reset? the same you did before? selling hope for months and taking the money while getting silent? you made me see the real ethics behind google

Actually, we reset quota across all plans. Everyone needs to build!

We heard concerns that Antigravity consumes many tokens for simple tasks now. So, we're adding Gemini 3.5 Flash (Low) as a way to optimize token usage for these tasks. In our internal testing, it generates around 45% fewer tokens than Gemini 3.5 Flash (Medium) and generally outperforms Gemini 3 Flash (High) on SWE tasks. We've also gone ahead and reset Gemini quota across all paid plans to make sure you have all the tokens needed to build for the next week 🙂

@elonmusk @iScienceLuvr @elonmusk @nikitabier have you guys consider not displaying the "username" and leaving only the @handle , maybe as optional view?

@iScienceLuvr Putting “Ph.D.” in your social media name is a sure sign of a pompous retard

This sort of mindset is probably why xAI failed to catch up to other frontier labs. If he wants to make SpaceXAI into a frontier lab, hope he changes his mindset. Though being a cloud provider is probably something they can easily excel in anyway lol (Colossus is impressive!)

qbit difussion sundays open.spotify.com/track/4DvhpE8O…

@DeryaTR_ nice sprites!

My AI coding hobby is recreating old games from the 1980s-90s, originally made for computers ZX Spectrum 48K, Commodore 64 & Amiga, which most youngsters may not know existed! Here is one called Exolon, the ZX Spectrum version, which I made with Codex as a browser game. Link👇

@zeroxjackson @Airaxora "don't eat the yellow snow"

@Airaxora greenp0ison iOS 5 jailbreak tutorial

pretend ur 11 and this a youtube search bar

@herkiparn @boyuan__zheng did you try to ask grok for a management template that suits both your project goal and working style?

Yeah, I used "own" too loosely there. I do not mean fork the CLI. API is probably the right way to do it. What I mean is the harness owning goal state around the CLI: objective, pending work, changed files, checks, pass/fail, next step. The colony sim is just one workload I use to stress that loop. It is not feeding prompts back into it.

One friction I keep hitting: the agent treats small edits as the end state. It makes one change, then I have to keep prompting "continue" to get back to the actual goal. A persistent goal system would help a lot here. Keep the high-level objective alive in the harness, even if the model cannot run forever yet. Then each edit is one step in the plan, not the finish line. Would make long-horizon agentic coding much more useful.

testing some grok guild-0.1 software refactoring, metrics got missing in the way but grok is solving toward the multi synth paradigm nicely and doing quite long runs! 10mins and 6mins last two prompts, such large work demands several hours of review and detailed examination to fine tune the scope, accompanying music is youtu.be/n7Pj3BMbPVo

@herkiparn @boyuan__zheng so are you kind of an "enhancing loop" between prompts by auto feeding from the colony sim stress test? btw you can test the cli via api, there is no need to "own" ;)

Haha yeah weird is a good description. Rust web engine, scripting runtime in C# , colony sim sitting on top to push-load test it. They all exercise the agent loops differently which explains call counts. About owning the CLI: definitely tempted but i'd rather have the harness figure out how to meet goal-state vs maintain my own fork for eternity. How does that cost scale at that volume? The context blowing up from compaction after a few rounds is what I would worry about on longer runs.

@patnrv @skcd42 is an outstanding model to me, i wonder what you do to experience such weakness, mind to share how?

@skcd42 Grok Build is the best TUI. Needs a better model. When is the 1.5T dropping?

Bug fixes shipping to Grok Build (release notes will be available in the TUI) 0.1.218 • Ctrl+X as default shortcuts help binding on Windows • Fix image pasting + shortcuts keybinding on Linux • User-specified duration for video generation tool • Ctrl+X as default shortcuts help binding on Windows • Support temporary screenshot images on mac • Validate image bytes to prevent retries • Improve compaction prompt (match training and rehydrate skills) • Increase ulimit on mac and linux, preventing ENOSPC errors from appearing and bricking the CLI • Multi-line image links are not clickable (and don't break anymore) We are focussing more on Linux and Windows so engineers on these platforms have a great experience.