Welcome to Electrified, it's your host, Dylan Loomis, quick shout out to my new or updated patrons, Douglas M and E3. Thank you for choosing to support the channel. Another thank you to the most recent two of you who decided to use my Tesla referral link to MythicC and Megan B. Congratulations on your new Tesla. And finally, in the spirit of gratitude, it's long overdue for me to take a moment to thank each and every one of you who have ever interacted with any of the sponsors of the channel or supported me on Patreon or X. To emphasize this point, Jordan from the limiting factor recently shared he crossed 10 million views over the past five years and he shared his ad revenue of about 83.6 thousand dollars over that five year period. We all know how much work Jordan puts in and he averaged over that five year period $1,500 per month from ad revenue alone. I wanted to share this to emphasize my point that all of the insilary support that you guys can offer outside of just watching and liking my videos goes a lot further than I think you may understand. Another way of saying it, if I didn't have channel sponsors and Patreon and a little bit on X, my wife would have a lot more questions about how I'm choosing to spend my time. So I mean it when I say had you guys not been so generous over the past few years in supporting electrified in those ways, the channel wouldn't exist as it does today. I know the sponsors can feel less than ideal but in this season of the channel they really are quite essential. I try to make them as educational and valuable as possible but seriously to anybody that has ever checked out any one of those companies I cannot thank you enough. And Jay if you happen to see this I'll say it again, congratulations my friend we all greatly appreciate all of your work.
Over the weekend Tesla posted produced our 100 millionth 4680 sell across all of our factories. Let's make some sense of this number. Now I know a lot of people like to look at pretty graphs so they can find the S curve in the data but in my opinion those can easily be manipulated to look any type of way. So for me personally I like sticking with the raw data. Running through this table we have the date that the announcement was made by Tesla, the actual milestone for the number of sells produced, then we have the days between each announcement. For example for Tesla going from 1 million cells produced to 10 million it was 484 days which is also 16.1 months. Then we have the additional cells produced during that time period so this would mean Tesla produced an additional 9 million cells during that 16.1 month period. Then we have the cells per month run rate based on that time period of data and here we have the number of Cybertruck packs per month that Tesla could produce given the current cell run rate. And finally I plugged in a formula assuming 95 watt hours per 4680 cell. So this number is the annual gigawatt hour run rate again for the snapshot of time between each announcement. Now yes I'm fully aware that back in 2023 and prior Tesla's 4680s were not around 95 watt hours per cell but they are today and I wanted this number to be uniform going back through each announcement.
The first thing to point out, this announcement Tesla called out across all factories whereas with the prior three announcements every time they said specifically at Gigatexas. And then the first milestone back in 2022 there was no call out but the photo was from Kato Road. And obviously Gigatexas was just opening in the summer of 2022. I totally understand if you want to argue these numbers should be a bit lower removing for any cells produced at Kato Road but personally we don't know for sure and honestly I don't really care where the cells are coming from because they're being made one way or another. And a few years from now 10 million cells here and there is not going to make that big of a difference anyway. It's been 101 days since we got the last announcement which is 3.4 months so in that time Tesla has produced an additional 50 million cells meaning during that time they're making about 14.9 million cells per month which would be enough for about 11,050 Cybertruck packs per month. We know that right now Tesla is likely selling around 15,000 Cybertrucks per quarter so the current run rate for 4680s is indeed well ahead of the Cybertruck ramp. And again assuming roughly 95 watt hours per 4680 cell the annual gigawatt hour run rate for Tesla during this snapshot in time is 16.9 gigawatt hours per year which clearly is a massive update from the 4.3 gigawatt hour annual run rate that we were dealing with with the last announcement.
Just to quickly zoom out I do have to call out that at Tesla's battery day they were expecting to be producing 100 gigawatt hours per year by 2022. So have they fallen woefully short of those expectations? Yes they have but are they now starting to make real exciting progress? The answer is also yes. A quick aside that's why personally I wish Tesla would stop putting timelines on things because that just removes one attack vector for critics to complain about the company. Of course as an investor and for the channel I love Tesla being forthcoming with timelines but if I had to choose I'd be fine without them. But check this out in just over 3 months Tesla has nearly quadrupled the gigawatt hour run rate. Further in my handy notes Tesla is retooling for the Gen 2 cell in quarter 4 of this year at Gigat Texas. They're adding 4 more lines which will be in addition to the first 4 lines. Those are to be installed starting in quarter 3 of this year and they've been under construction since quarter 3 of last year. This means this increase in cell production was largely from the phase 1 or the first 4 lines and we have another 4 set to come online by the end of this year.
The last update we got about Giganavada and Tesla and Panasonic making 2170 cells was that their annual capacity is between 35 and 40 gigawatt hours. Thus Tesla's captive 4680 production largely from Gigat Texas and some from Kato Road is nearly half of that number. And Tesla and Panasonic at Giganavada have been working on those lines for nearly a decade. For more context Toyota is setting up a factory with Panasonic set to open in 2025 and the annual capacity of that one is set to be 30 gigawatt hours. You can see in terms of a run rate Tesla added over 12 gigawatt hours over the past 3.4 months so assuming no S curve and just linear growth even if they were to add another 12 gigawatt hours for each 6 month period that would be 24 over the next year which would hypothetically put them at roughly a 40 gigawatt hour annual run rate. If it's not clear this is a big deal because those numbers don't even factor in Tesla's plans to start producing full dry battery electrodes for the anode and the cathode. Hopefully they start that by the end of this year.
Recently we highlighted that the energy density of Tesla's 4680 is now on par with much of the competition for nickel base cells and they're clearly now hitting some legitimate scale figures starting to ramp up that S curve which over time will bring down the cost. Earlier this year who knows if the report is actually true but the word was that Elon told the team if they don't address certain issues preventing production from being scaled up before the end of this year the project would be abandoned. The project being 4680s. Pair that with a tone on recent conference calls being that yes Tesla is making progress with the 4680s but they're still going to be heavily reliant on their suppliers which is definitely still the case but I think there's a case to be made that over the past year Tesla has been sandbagging a bit when it comes to the 4680 status.
Last thing I'll say because I could easily drag this out again Tesla is expected to deliver around 15 000 Cybertrucks per quarter right now and given this most recent snapshot their monthly run rate of Cybertruck packs is in the neighborhood of over 33 000. So hypothetically Tesla could double Cybertruck deliveries overnight and they would still be making enough 4680 cells. Anything can happen but if I look forward two three four years and Tesla does make full dbe cells and they add some silicon and they continue to add lines there's a real case to be made that Tesla could be one of the premier battery manufacturers on the planet. Not yet in terms of overall gigawatt hours per year but in terms of energy density and the cost per cell. I really wanted to emphasize this point because it feels like over the past year a lot of people including Wall Street have begun brushing off the 4680 project and writing it off as more of a side project for Tesla that may never really materialize into anything like they thought it could be. But in my opinion with this update and analyzing the numbers I absolutely think that narrative should be changing.
Tesla scopes shared a new Tesla software update 2024.27.10 this one does have fsd 12.5.2.1 I just got it on my hardware for vehicle today. The word is this is also going out to hardware three vehicles but this one does not yet have actually smart summon. The Department of Energy put out an infographic highlighting that a typical EV is 87 to 91 percent efficient compared to 30 percent for a conventional gas vehicle. The figures for EVs is indeed after taking regenerative breaking into account. I think most of us here already understand this principle but I wanted to let you know I'll have this linked below in case you need a resource to share with family and friends.
If you go to the Tesla careers page and search Philippines you will find a few different job postings specifically in to gig. There are 15 positions listed as of today and one of them is indeed store manager. It appears as though Tesla is set to make its official debut into the Philippines there are some Teslas in the area but they've all been imported privately. Tesla is likely setting up a physical location for vehicle sales in the next few months. The EV market in the Philippines is absolutely still in its infancy they need a large build out of infrastructure but Philippine auto sales for 2023 were over 441,000 units. 75% of those were commercial and 25% were passenger. For last year Toyota held an astounding 45% market share. The top luxury brand last year was Lexus selling 1,843 units. If Tesla can split that difference over the next few years and eventually hit a 10% market share that would be an additional 40,000 units per year. This is from yawn in the EV universe newsletter he was highlighting in Canada we got the results for the first half of this year.
Just over 84,000 EVs were sold during that time period up 41% from the first half of 2023 and full EVs now have a 9.1% market share of all cars sold up from 7.2% over the first half last year. And how about Hong Kong over the same time period they had just over 20,000 EV sales which is up 51% year over year and full EVs now make up over 83% of the overall auto market share. So Norway has a bit of competition. Just two quick data points to further dispel the myth that full EV sales are dying. Cybertruck Engineer Wes shared a post on X talking about Conway's Law and his experience at Tesla.
Basically Conway's Law he said is organizations which design systems are constrained to produce designs which are copies of the communication structures of these organizations. Translation the more silos you have across company and separate divisions the less efficient your overall end product and design will ultimately be. He shared a story about how Tesla ended up moving to the structural pack but we've talked about that in the past so pause the screen if you'd like to read. I just wanted to share he said someone just needs to question if there's a better solution in a team open to criticism. This mindset to work together to make the best product regardless of ego is where you end up with the most innovative products.
He then went on to list a few examples in the Cybertruck specifically and concluded saying the excitement and motivation by everyone involved to work across boundaries and actively break down Conway's Law is one of the many reasons I love working at Tesla. It's this right here which is why I've often said even if these other companies hire great AI and engineering talent and even if it's from Tesla that does not mean that the culture and the mindset of the organizations at large at these other companies will be able to immediately mimic what Tesla has been doing. In the comments West said there are so many examples because it's ingrained as a mindset in all the teams and encouraged by our leadership.
That is a culture that's built into the Tesla ethos that cannot be duplicated no matter what in a short amount of time at other companies. That's part of why Jim Farley has said this year that their skunk works project in California is effectively an entirely different company or subsidiary under Ford. He didn't want that team to be burdened by all of the structures that have been put in place at Ford over the past few decades. It's true many of these OEMs have finally realized that they have to make drastic changes to their business model to ever make EVs that are attractive and that they can make a profit on. But to truly have an agile manufacturing and engineering company where every employee is fully bought in and trained to think like this that's going to take years and years if not decades.
The Tesla semi is officially set up and ready to go for the IAA event that has started today and Dan Priestley will be giving his keynote tomorrow on Tuesday. Some people from the Tesla community like yawn will be there going to the event so we're likely to get some nice details in the next two or three days. Larry Ellison, a close friend of Elon, had some interesting comments about AI. In Nobu Palo Alto, I went to dinner with Elon Musk, Jensen Wong and I went to dinner and I would describe the dinner as oracle and me and Elon begging Jensen for GPUs. Please take our money. Please take our money. Take no, no, take more of it. You're not taking enough that we need you to take more of our money. Please go. He said it worked, which is encouraging, but for Elon is that for Tesla or XAI. I'm not sure. Has anything like this ever happened before?
You know what basic stakes is? You know what it costs to build a frontier model? Anyone know over in that's three, three years? Yeah, a hundred billion. That's kind of gets you in the game. A frontier AI model like he said is really just a cutting edge model at the time. In a few years, the entry point just to get in the game could be a hundred billion dollars. I would add that's one more reason for the camp that argues Tesla should not do a stock buyback because they may need a lot more capital than I think people are expecting right now just for AI compute. James Stevenson shared a great chart highlighting net income as a percent of revenue comparing Tesla to BYD. Dating back to 2017, the high watermark metric for BYD was five percent. Whereas for Tesla last year, this number hit 17 percent. More importantly though, you can see the trend line for Tesla is steadily increasing, whereas the trend line for BYD is largely staying flat. I did ask James if he thinks BYD is unprofitable without subsidies because the internet in articles will give conflicting information. He said I don't follow them closely enough to understand the magnitude of their subsidies, but his intuition is the Chinese government has prioritized goals other than BYD shareholder wealth maximization. Translation BYD is really focused on unit growth and maximizing economies of scale.
Monroe and the team have a new business line. Apparently in partnership with Armin, they have started their own business designed specifically for Tesla vehicles. They now have their own lineup of wheels that are lighter, stronger, and more aerodynamic than anything else on the aftermarket. On their website, they currently have options for the Model 3, the Model Y, and the Cybertruck. The latter of which they're saying is bulletproof. All of these start at over $3,700 for a set of four. Wanted to get that on your radar, it's Monroe Delta Wheels dot com.
I've seen some folks saying there are now more EVs on Norway's roads than petrol vehicles. This in terms of the overall fleet. However, that data point conveniently leaves out the roughly 1 million diesel cars that are on Norway's roads which are in addition to the number of petrol cars. So yes, by the end of this month, there will be more EVs than just petrol alone, but that excludes diesel. Separately I found with almost 370,000 more diesel cars than BEVs, analysts estimate it'll take three to four years for BEVs to overtake diesel vehicles as well.
We got an update on the Tesla diner from 24-7 Tesla and they're currently doing some grading for the lot before putting down the concrete or the pavement. He also highlighted all of the LED lights strips that are being added to the outside of the building. That will be pretty cool at night, and he said for the interior, they still haven't made too much progress. But for the interior, he said they're waiting until they can close up the exterior and entirety before starting that.
Tesla owners Mexico shared that the first Cybertruck deliveries are about to begin in Mexico. And the Cybertruck program manager said that Cybertruck deliveries to Canada are likely to begin late October. Tesla just submitted plans to install a sculpture at the southwest corner of their engineering HQ in California. The concept is a stylized, low-poly California grizzly bear meant to combine Tesla's design language with California's bear and highlight the common message of strength and unyielding resistance. The bear would be 16 feet in 5 inches long and 11 feet in 5 inches tall. Tesla is also preparing to install glowing spotlights around the bear. The RGB lights would allow Tesla to control luminosity, intensity, temperature, and color and allow the company to change the lighting based on the season or special events and holidays.
A little tidbit for FSD in China, Tesla has now added FSD to the user interface, but the option is still grayed out. However, it is now finally visible. Bloomberg highlighted it took BMW more than two years to discover the extent of a braking system fault that's expected to cost the company $1.1 billion. Customers began complaining about faulty brakes back in June 2022, but it wasn't until last month when BMW realized as many as 1.5 million cars might be impacted. It's pretty wild that a software update for Tesla to change the icon size by a millimeter is going to get more attention than a billion dollar brake fix from BMW.
Tesla stock closed today at $226.78 down 1.52% while the Nasdaq was down 0.52%. It was a low volume day for Tesla trading about 21 million shares below the average volume in the past 30 days. Hope you guys have a wonderful day, please like the video if you did. You can find me on X-linked below and a huge thank you to all of my Patreon supporters.