RT Tesla Owners Silicon Valley (@teslaownersSV)In depth look at the Tesla CybercabThis gives us the clearest view so far of the refined interior and exterior tweaks:• Frameless windows for a cleaner look• Updated front bumper with a longer lower splitter• Minor revisions to the rear hatch• Full production-style front lightbar that integrates the headlights• Fresh wheel covers• Redesigned dashboard and door panels inside• New carpeting material throughout the cabin• Refined seat cushions and revised center c ...

Highest precision on DeepCodeBench.When it comes to understanding, generating, and reasoning real-world code, Grok 4.1 outperforms Gemini Pro 3 and GPT-5.1.Grok is ahead; this is measurable performance, not hype. https://t.co/CP2djwhRxa ...

Grok has the highest retrieval score on LegalBench, surpassing all of its main rivals in contract analysis and sophisticated legal reasoning. No other model can match this benchmark, the difference is evident, and the outcomes are quantifiable. https://t.co/AMsr8jjqml ...

RT Tesla Owners Silicon Valley (@teslaownersSV)Just took a fun little test drive in Los Gatos on FSD 14.2.2: house → Apple → McDonald’s → Starbucks → back home.Honestly the Tesla Autopilot team is absolutely cooking right now 🔥The highway portions were butter, surface streets felt confident, lane changes were smooth, unprotected lefts were clean, and it handled the weird Los Gatos traffic circle like it’s been there a hundred times.Some small notes though (because it’s still improving):• Apple parking lot w ...

BREAKING: Grok is officially part of X Chat as of today. Long-press any message → select “Ask Grok” and you’re good to go. https://t.co/t5QMt6nOFr ...

RT Tesla Owners Silicon Valley (@teslaownersSV)Just took a fun little test drive in Los Gatos on FSD 14.2.2: house → Apple → McDonald’s → Starbucks → back home.Honestly the Tesla Autopilot team is absolutely cooking right now 🔥The highway portions were butter, surface streets felt confident, lane changes were smooth, unprotected lefts were clean, and it handled the weird Los Gatos traffic circle like it’s been there a hundred times.Some small notes though (because it’s still improving):• Apple parking lot w ...

Grok 4.1 outperforms Gemini Pro 3 and GPT-5.1 on real-world codebases, APIs, and complex algorithms with an end-to-end accuracy of 85.6% on DeepCodeBench.designed with developers in mind. tested in real-world situations. https://t.co/G7SzCxNtyu ...

With a 96.1% retrieval score, xAI's Collections + Grok 4.1 outperforms Google and OpenAI on tabular and numerical data from intricate documents such as SEC filings.The edge is hybrid search. The key to success is accuracy. https://t.co/4FqCKMb1wW ...

Grok is demonstrating its superiority in actual RAG performance 🚀Highlights of the benchmark:• Coding: 86 (best overall)• Finance: 93.0 (clear lead)• Lawful: Competitive with leading modelsGrok 4.1 from xAI When it comes to lengthy, intricate documents, where precision is crucial, Fast either matches or surpasses industry leaders. ...

Just one month after surpassing 8 million active customers, Starlink has now reached 9 million across 155 countries.The map keeps getting more bizarre, showing off-road rigs in remote areas, desert camps in Mongolia, frozen lakesides in Japan, and thatched roofs in rural Africa.The speed at which this is moving is astounding. ...