MacBook Pro M5 Max: What Apple Is Really Building

The MacBook Pro M5 Max, M5 Pro are Apple's most powerful laptops to date, but they may be one of the most consequential laptops since the M1 MacBook Pro. Why AI? So, let's talk about the [music] specs on these machines real quick here. On the M5 Max MacBook Pro, got 18 core CPU, 40 core GPU, 128 GB of unified memory, 8 TB of storage, nanoexture display. This is the creme de la creme. This is an unbelievable laptop. Now, we move over here to the Mac Studio M3 Ultra, which I'll be comparing it later on in this video. You got 32 core CPU, 80 core GPU, 512 GB unified memory, and 8 TB of storage. Yeah, there's a lot of computing power on this desk. We'll talk about real world usage. which one should you get later on, but this is just for the test to see what these machines can do against each other. Okay, M5 Max, M3 Ultra, and I'll be honest with you, some of the results you're going to see are pretty freaking amazing. All right, when we talk about the world of AI, right, there's been a lot of criticisms towards Apple in terms of their their direction for AI. Why are they behind the Googles, the Chad GPTs, the clouds? where is their AI solution at? What they've been doing is making hardware powerful enough, and that's why this new architecture in the M5 Max and M5 Pro chipset, even the M5, is different than the others, is to take advantage of these larger number parameter local language models and use AI on a laptop where you never could before. at these sizes. You can actually take all your data, all your information, if you're a content creator, all the transcripts of your videos, put it into a filing system and your AI can access this. It can give you all the relevant information to help you create even better content. Now, you can take this MacBook Pro M5 Max and it has the power of an AI workstation in a device like this. This is fascinating stuff because the potential to have your own personal assistant that knows everything about you that you don't have to pay a monthly subscription for that you can constantly feed it information and don't worry about it disappearing is now possible for all of us that can get into this system. Now, it doesn't have to be the M5 Max. It doesn't have to be the M5 Pro. Even the M5 has more AI capabilities to it. That's what makes this exciting. So, I've done some tests here. I'm using LM Studio. Okay. Now I'm using a uh Llama 3.3 70 billion parameter. Okay. 6bit local language model. And from there I actually took a 45minute transcript. It was an interview I did with the CEO of Sigma Lenses and cameras. I took that and then I fed it a very large prompt. Various different steps I wanted to do in terms of getting the right information. So I wanted it to identify the five most compelling moments from the conversation. Number one, turn those moments into short form video ideas suitable for Instagram res or YouTube shorts. Then write three strong YouTube titles for a long form video covering this interview. Suggest three thumbnail concepts that would attract viewers interested in photography and camera gear and provide a short hook, two to three sentences that could be used uh as the opening of the video on the M5 Max. Okay, it took 2 minutes and 52 seconds to do that entire large prompt analysis from start to end. Now, I also tested this with my maxed out Mac Studio M3 Ultra that came in at 1 minute and 21 seconds. Now, you're thinking, of course, this is a Mac Studio. This is a MacBook Pro M5 Max. Even though this is the latest chipset, this is going to be a little bit stronger when it comes to single tasks. Keep that in mind. single tasks. It gets exciting in this video. So then after that, I came up with a second prompt. Very short prompt there uh to it and basically just asking it to find out some find out a topic that would be great for photographers and videographers. Very simple, like some some information that would be great for them to know. That took about 34 seconds on the MacBook Pro M5 Max and 37 seconds on the Mac Studio M3 Ultra. They're pretty much neck and neck when it comes to more shorter prompts. Okay. Then I used an application that's free to download. It's called Draw Things. Now this is where you can actually take um local image generation models and I used Flux 2 for this one. And I actually developed various different sizes of images with a prompt and I'll put that in my description so you can actually play with this at home. The M5 uh for 515 x 515 image generation, which is more of a smaller image. The MacBook Pro M5 Max came in at 1 minute and 32 seconds. Not bad. The Mac Studio M3 Ultra 1 minute and 48 seconds. Interesting, right? This gets into the architecture about the M5 Max, which we'll talk about later on. Then I went to a 1024x 1024 image. The MBook Pro M5 Max 5 minutes 15 seconds. Mac Studio M3 Ultra 5 minutes 30 seconds. So, it's a little bit behind. Then I decided to make a large image at 2048 by 20048. The MacBook Pro M5 Max came in at 15 minutes and 53 seconds. It's a much larger image, but the Mac Studio M3 Ultra came in at 25 minutes. That's where the architecture, the 40 core GPU, the 18 core uh CPU, all this coming into play and this fusion design that's inside of this really helps with image generation. Apple even talks about being, you know, 1.4 four times faster than even the M4 Max. And you're even seeing it here in the M5 Max at how powerful this is. I decided to use Da Vinci Resolve since we all use Resolve, right? So, I've got some R3D footage from a friend of mine, Peter Choy, has a Red Vaptor 8K dropped into the timeline. Okay, it was a 4K timeline. The video length was about 2 minutes 15 seconds. Did some color grading, did a couple effects, very basic, not too much, but just enough to see how this would perform. So, and I was converting it down to H.264 264. With that, Da Vinci Resolve, the MacBook Pro M5 Max took 1 minute and 58 seconds. The Mac Studio 1 minute and 27 seconds. So, the Mac Studio pulls ahead, rightfully so. This is where the wheelhouse is going to be for something like this. Then, I decided to do a very big test. What if I ran all three models, image generation, local language model, and resolve at the same time on the M3 Ultra and on the MacBook Pro M5 Max pushing these things. And you can see it from the graphics as you've been seeing out throughout this video in terms of the the recordings of the screen. It's pushing the GPU. It's pushing the CPUs. You see a lot of power coming into these, especially when you're doing all three at the same time. I will tell you this, none of them crashed. But let's look at the numbers here for a second. The Mac Studio M3 Ultra with LM Studio to do the same prompt that I talked about earlier, I identifying the five most compelling moments in the conversation, etc., etc. Took 8 minutes and 5 seconds. Draw things took 7 minutes and 58 seconds. And the resolve render also took exactly 7 minutes and 58 seconds. But what about the MacBook Pro M5 Max for LM Studio? Took 6 minutes and 22 seconds, almost 2 minutes faster than the M3 Ultra. draw things which was with the flux 2 image generation 1024 x 1024 took 5 minutes and 3 seconds and the resolve render took 6 minutes and 43 seconds versus 7 minutes and 58 seconds. So almost a little over a minute faster. That's with this laptop. By the way, it wasn't plugged in running purely off the battery. What was the battery drain like since I didn't plug it in? Maybe 10%. Yeah, I was pretty blown away by that. I'm like, damn, I wasn't expecting that. But this is just to give you guys an idea of what this is capable of that you can run a local language model, 70 billion parameters, llama 3.3, 6bit, very high. It's a very big language model to download. I think it's one of the largest you can download right now. 70 billion is pretty much where it's at. Then you can do image generation 1024 x 1024 with a very detailed prompt and then render 8K footage on a 4K timeline to H.264. A lot of compression going into there and still get faster than the M3 Ultra. But the M3 Ultra does excel on single tasks as we saw with the local language model. But when it comes to image generation, this is faster. And that's where we're seeing the evolution of this chipset from Apple and why it's so important is because now we literally have an AI workstation in a laptop. This is a beginning of a future where our personal assistant is with us wherever we go. Doesn't have to connect to the internet. It's private. It's secure. It's with us. And that's what makes this laptop so special and these chipsets so special. With that said, do you need it? Let's talk about that for a second. What I got on my desk is probably overkill for 95 to 96% of people out there. If you're into photography, take out your Leica, your Sony, your Nikon camera, taking some photos, do you need to go for the Max? No. You could actually get away with a MacBook Air M5 and be okay. M5 Max is really going to be for those who are again getting into AI, you know, local language models, image generation as well, what you saw from some of these tests, you know, really pushing it when it comes to, let's say, Resolve and Final Cut. That's where the Max is definitely going to come into play. You can do a lot of that on the Pro, too, but the Max will just give you a little bit more headroom. But the M5 Pro is a phenomenal machine for most of you out there, and the M5 Air is awesome. Absolutely awesome. Anyway, guys, that's kind of it. It's kind of a say first impressions because I've only had it for a couple days, but damn, this is good, right? Thought this would be great for you guys to see how powerful this machine is and the potential what this has for the future. Not only for content creators, AI development, but also businesses that want to have their own AI. Now, they can do it on a laptop. That's pretty freaking insane. Anyway guys, that's it. Like, subscribe. Thank you for the support. Greatly appreciate it. And I'll chat to you all very, very soon.

#apple #MacBookProM5Max #M5Max #MacStudio The new M5 Max MacBook Pro is more than just a spec bump; it's a portable AI data center. In this video, we put the maxed-out 128GB M5 Max head-to-head against the ultimate desktop powerhouse, the Mac Studio M3 Ultra.Instead of relying on cloud subscriptions like ChatGPT or Gemini, Apple's new Fusion architecture and Neural Accelerators allow creators to run massive AI models 100% locally and privately.To see what this 40-Core GPU is really capable of, I ran the "Trinity Stress Test." We pushed a 70-Billion parameter Llama 3.3 model in LM Studio, generated extreme high-res 2048x2048 images using FLUX.2 [dev] in Draw Things, and rendered 8K RED V-Raptor footage in DaVinci Resolve—all at the exact same time, while running purely on battery power.The results will surprise you. Image Generation Prompt For FLUX.2 A hyper-realistic scene of a futuristic city at night, with neon lights reflecting off wet streets. A detailed character stands in the foreground wearing a chrome visor, with intricate futuristic armor reflecting the city’s lights. Drones fly in the sky, and a massive digital billboard displays an alien language. The scene is cinematic, with volumetric fog and dynamic lighting. Chapter Markers 00:00 - The Most Consequential Mac Since M1 01:12 - Your Private AI Workstation 02:38 - Pushing a 70B LLM Locally (LM Studio) 04:28 - Next-Level Image Generation (FLUX.2) 05:52 - Crushing 8K RED Footage (DaVinci Resolve) 06:33 - The Ultimate Multi-Tasking Flex 08:13 - Mind-Blowing Battery Performance 09:40 - Which M5 is Right For You?