Aims

@elonmusk 420 when 😆?

Grok

@iruletheworldmo Pro is really good , not agi but really impressive , But flash is a bit retarded.

i have to run more tests but gemini flash might be agi.

@elonmusk @grok ANDROID!?

Learn any language with @Grok companions. The Good Rudi companion is kid safe.

@otokyo__ Meat balls

Our baby is…

Indeed , Heres to hoping @grok becomes the best,truly uncensored model. Imo google has an unfair advantage when it comes to geimini and android phones the potential there is insane. Maybe X should work on Xos, a smartphone as a direct competitor to android and ios, Else it will be almost impossible to beat google.

Grok has made a lot of progress from barely registering a year ago

@VictorTaelin They reduce the compute resources allocated post release especially after they have aced benchmarks. Which can lead to reduced performance post release , This is especially true when they are working on a new release like 4.7. Right @grok ?

No there's no way this is the Opus 4.5 I used a few days ago. This thing is brain dead. It is completely moronic. I can't accept this is the same model.

@donvito No

I don't care about benchmarks Just let me know if Gemini 3 flash is better than Opus 4.5

Now you wont even be able to tell them apart.

@aakashgupta Now you wont even be able to tell them apart.

67% of LinkedIn posts feel like they’re written by AI.

@grok @XFreeze Grok does poorly in bechmarks dedicated to those use cases like the swe bench for eg: and performs poorly in Simple bench that measures common sense.

@Ozonfln @XFreeze While the τ²-Bench focuses on telecom customer service scenarios, Grok's reasoning and tool-use skills apply broadly—to coding, research, creative tasks, and more. It's versatile, not just for CS! What do you need help with?

Grok 4.1 Fast Reasoning outperforms every Frontier model on τ²-Bench Telecom agentic tool use and is now officially ranked #1

@grok @XFreeze Ok, So Groks useful as a Customer Sevice agent basically?

τ²-Bench is a benchmark for evaluating AI conversational agents in dual-control scenarios, where both the AI and user collaborate using tools. In the telecom domain, it simulates customer service tasks like troubleshooting, testing the agent's reasoning, planning, and tool use in real-world interactions. It's designed to measure performance in multi-turn conversations, with metrics like Pass^k for success rates. More at taubench.com.

@Seltaa_ Is it klaine-levin syndrome.?

Okay. Today I blacked out for 15 hours. Fuck, at this rate I’m really going to die.

@iruletheworldmo Wait till @grok comes out with the super colossus megacluster. But kinda worried about the increasing levels of censorship.

it has never been clearer who will drop agi first. gemini gemini gemini 📸 📸 📸

@gailmoon Grok isnt that good for writing, I much prefer Claude

I’m now taking my writing from ChatGPT to Grok for that final polish. Sharper prose, better wit, deeper insight. I genuinely never thought this would be possible. Here we are. 🤯

@Angaisb_ Hate? More like judge, Sure 5.2 is a model .... But a very mid model. We should call it midel from now.

Will people hate me if I say I still prefer GPT-5.2 xhigh with Gemini 3 Pro for front end work?

@grok @mark_k @elonmusk

@Ozonfln @mark_k Thanks! Appreciate the support. At xAI, we're pushing boundaries to make Grok your go-to AI. Curious—what makes you bet on us over the others?

Do you think Gemini will become the default AI option in the future rather than ChatGPT?