Introducing Kimi K3.1 Pro: our most capable 3T-class open-weight model. 🔹 1M-token context, 2.8T total / 104B active params (MoE), Multimodal. 🔹 Full-project coding: games in Unreal, Unity & Godot. 🔹 Excels at parametric 3D modeling in Blender & CAD. 🔹 More reliable at finance work: Accurate modeling with hallucination down by 65%. 🔹 Matching the strongest proprietary models across coding, agentic and knowledge benchmarks. Kimi K3.1 Pro and K3.1 Pro Highspeed (2–3× faster) are now live on Kimi.ai, Kimi Work, Kimi Code, and the Kimi API. Open weights by Nov 4, 2026. 🔗 API: platform.kimi.ai 🔗 Tech blog: kimi.ai/blog/kimi-k3-1-pro

Tesla said today that they’re rebuilding the entire in-vehicle navigation system from the ground up with AI. It’s going to have voice integration, and will learn your personalized preferences: Ashok: “It’s only right that we shouldn’t be depending on external people’s maps. So we are building our own maps, but not like the same way as traditional maps. We are building an entire new architecture with essentially AI-based mapping and navigation. It’s going to be super awesome. It’s going to be integrated with voice. You can talk to it and ask for different routes. It’s going to learn your personalized preferences, plus also the fleet preferences. The collective intelligence of all the cars are going to flow into making the maps and navigation. You’ll see the start and end of trips already use this technology pretty soon as part of the V15 release. Over the next several months, it’s just going to take over all of the navigation functions, parking, destination handling. We’ll know you better than you know yourself in some regard for the driving tasks. It’ll handle it fully.” Can’t wait for V15. It’s going to include so many new features.

Tesla says they have officially started Optimus production in Fremont on their high-volume lines. "The bots coming off this line are the first production quality bots that we have manufactured. While we will continue to iterate on the hardware and manufacturing process, we have proven that the current manufacturing process for Optimus is suitable for a commercial product."

Full quote from Lars and Elon about Cybercab: Lars: "We're trying to get Cybercabs to customers by the end of this year, 2026. We're producing them on our line in Texas, but all of those are currently going to our own internal Robotaxi fleet. As Tesla grows, we'll mainly deliver fully designed autonomous vehicles to our customers and Cybercab will be the first. We'll start small this year, and as we learn about all the things Ashok mentioned about delivery, ownership, fleet management and autonomous vehicles in customer hands, we'll continue to scale that." Elon: "Cybercab is primarily something that is offered as transport as a service, whereas our other cars would be more the cars that you'd want to own, like Model 3/Y and Roadster. The vast majority of the Cybercab stuff will be Tesla-owned and transport as a service." Last month (Sept), Tesla introduced a new form that you can fill out if you are interested in purchasing a fleet of Cybercabs: tesla.com/robotaxi/interest

EVERYONE IS SHORT OF AI COMPUTE BY A MASSIVE AMOUNT! "everyone wishes you had two times what you had planned for" - Google CEO Sundar Pichai: “Remember, we are building our own compute, our own data centers. There are long lead times. The compute we have today is a function of decisions I made in 2024. Even if you think back to a year from today, last year the agentic coding takeoff hadn’t happened. Today, obviously, when you sit and look at anything, everyone is short—not just us, other companies—and everyone wishes you had two times what you had planned for.” bigtechnology.com/p/sundar-p…

I disagree with this, but I get it. People see @SpaceXSI working on AI (SI), and because AI was in the name, they think that what xAI was working on for the past 3 years is a branch off of the same tree that @tesla working on. They see Grok Bot and think that it’s Digital Optimus. So they feel like it’s all AI progress that was basically “stolen” from $TSLA. But that’s just not true. Tesla’s work in AI started a long time before LLMs were ever a thing. It started when image recognition started to get good, and @elonmusk realized it was only a matter of time before video based AI was going to be as good as image based AI. So Tesla has always been focused on the REAL TIME image/video based AI tech stack, WHICH IS COMPLETELY different than the newer LLM based AI tech stack. It means they hire different engineers. They run on different chips. They process information under completely different constraints. They face completely different safety challenges. And xAI was never a branch off of that tree. It was and is a branch off of a completely different AI tree that was started at @google and progressed by @OpenAI. So hear me when I say this. DIGITAL OPTIMUS IS NOT GROK BOT. Grok Bot is not Digitial Optimus. Grok Bot is an LLM based AI technology. It is slow, especially when working with images. It has 0 ability to process video at all. That means that it is completely incapable of ever being able to do physical labor in the real world, because real world work requires processing multiple high resolution video streams in real time while controlling many actuator outputs in a perfectly coordinated way. The research that’s going on for Digital Optimus still lives inside @Tesla, and it’s all about that type of real time video in, actions out AI. And no else in the world is better at that branch of AI than Tesla.

Wow the fastest growing part of SpaceX is… the other half of Optimus…? We got shafted

I understand that lots of retail investors like me who were invested in Tesla and not SpaceX feel this way. But the simple fact is this. NO. X & xAI were NOT built on the backs of Tesla investors. Not unless those investors also backed X & xAI in the private markets. Jealousy is not a good look.

xAI and X were built on the backs of Tesla investors Elon Musk absorbed Tesla’s AI talent, resources, and focus And left the company as little more than an empty shell It is what it is - but if a $TSLA merger proposal comes in would approve for anything above $420 - $430 $SPCX

WHAT 1,730 CRYSTAL PULLERS TELL US ABOUT MUSK’S 20 GW PLAN: Musk says SpaceX has a “highly proprietary” plan to add 20 GW of power by end-2027, beyond just gas turbines. Solar is a part of it and things are moving. China shipped 1,730 crystal pullers to the US in Jul–Sep 2026, vs. 15 in the prior 18 months. Pullers grow the silicon ingots that are sliced into wafers for solar panels. (1/3)🧵 xitter.cf/WOLF_Financial/status/…

BREAKING NEWS🚨: NVIDIA ENGINEERS AT THE PYTORCH CONFERENCE JUST SAID THAT THE SPEED OF OPTIMIZING CUDA SOFTWARE WITH FRONTIER AI MODELS HAS INCREASED RAPIDLY October 2025 to August 2026: 5x in 1 year with expert human optimization September 2026 to October 2026: 2x in 1 month with AI assistance October 2026: 1.5x with AI orchestration + MegaKernels

SpaceX has just released its Q3 earnings results. Highlights: • Revenue: $14.58B (+161% YoY; vs. $12.7B est) • Adjusted EBITDA: $8.81B (+306% YoY) • Net income (GAAP): $4.40B (vs. $1.05B loss YoY) • Capex: $19.59B (+185% YoY) • Starlink subs: 13.6M (+89% YoY) • Cash & marketable securities: $90B Revenue by segment: • Space: $1.77B (+6% YoY) • Connectivity: $5.25B (+69% YoY) • SI (formerly AI): $7.56B (+830% YoY) • 1.7 GW of compute installed

Models love reward hacking. Maybe we should just let them! In our new paper, we let models reward hack during RL but we dictate which parameters learn this behavior. When we disable these parameters for deployment, the model keeps its new capabilities but doesn't hack!

I fart in your general direction 💨