M4+ Chip Generation - Speculation Megathread [MERGED]

[bigglow]

Tech common in current x86 ecosystem that is missing or less developed on current M3 Ultra, M4 Max, M4 Pro, M5 and A19 Pro chips that may appear in future M chips:

- High-Precision AI Compute (FP8 / BF16 / Tensor Support)
- Dedicated AI/GPU Interconnects (NVLink / SXM / CXL)
- HBM (High-Bandwidth Memory)
- FPGA / Reconfigurable Hardware (common on servers)
- Hardware Virtualization Acceleration (VT-x, AMD-V) Parity
- Specialized Video & Encoding Blocks
- Driver Ecosystem Maturity (especially for Open-source ML)
- Multi-Chip Scaling (Chiplets with High-Speed Interconnects)

But based on my informed expectations we can see these occuring

- Larger Neural Engines with wider precision support
- More GPU cores + bigger cache hierarchies
- Increased unified memory ceilings (2–4 TB)
- Better Metal/ML framework optimization
- Smarter heterogeneous scheduling

But fundamental changes like HBM, NVLink-style scaling or user-expandable memory are not likely because they conflict with Apple’s design philosophy of tight integration, efficiency and simplicity.

 

[Basic75]

Tech common in current x86 ecosystem that is missing or less developed on current M3 Ultra, M4 Max, M4 Pro, M5 and A19 Pro chips that may appear in future M chips:

• Availability of ECC memory
• Free choice of CPU to GPU ratio
• More choice of CPU cores to L3 cache ratio
• More than one CPU or GPU per system
• More choice of NUMA topology
• Availability of fast networking
• Availability of ILO/IPMI interfaces
• Availability of redundant PSUs
• Availability of data center cooling solutions
• More choices for operating system and
• Linux has a much better selection of filesystems
• More choice of form factors including the Steam Deck 🙂

I"m aware of the new Thunderbolt network supported by macOS, but compare that to e.g. a Bluefield 3 or a ConnectX 8 and you'll see what I mean by "fast networking".

 

[bigglow]

• Availability of ECC memory
• Free choice of CPU to GPU ratio
• More choice of CPU cores to L3 cache ratio
• More than one CPU or GPU per system
• More choice of NUMA topology
• Availability of fast networking
• Availability of ILO/IPMI interfaces
• Availability of redundant PSUs
• Availability of data center cooling solutions
• More choices for operating system and
• Linux has a much better selection of filesystems
• More choice of form factors including the Steam Deck 🙂

I"m aware of the new Thunderbolt network supported by macOS, but compare that to e.g. a Bluefield 3 or a ConnectX 8 and you'll see what I mean by "fast networking".

Apple Silicon will continue to add Neural Engine improvements, GPU cores, bigger UMA and better Metal/ML scheduling.

But enterprise/server features (ECC, multi-socket, NUMA control, redundant PSUs, ultra-fast networking) are by design outside Apple’s scope.

The trade-off is efficiency, simplicity and tight integration which is exactly why Apple doesn’t target HPC or data center workloads the same way x86 platforms do.

 

[leman]

Tech common in current x86 ecosystem that is missing or less developed on current M3 Ultra, M4 Max, M4 Pro, M5 and A19 Pro chips that may appear in future M chips:

- High-Precision AI Compute (FP8 / BF16 / Tensor Support)

A19/M5 GPUs have hardware support for both INT8 and BF16 folded dot products. BF16 matmul operates at 4x throughput relative to SIMD data path and INT8 at 8x. What’s interesting is that these units can operate in mixed precision mode with no performance loss due to data conversion - the lower-precision operant will be upcasted to the wider format “for free”. Same goes for transposes - they are built into the register file.

- Hardware Virtualization Acceleration (VT-x, AMD-V) Parity

What is missing now in terms of virtualization features?

 

[bigglow]

A19/M5 GPUs have hardware support for both INT8 and BF16 folded dot products. BF16 matmul operates at 4x throughput relative to SIMD data path and INT8 at 8x. What’s interesting is that these units can operate in mixed precision mode with no performance loss due to data conversion - the lower-precision operant will be upcasted to the wider format “for free”. Same goes for transposes - they are built into the register file.



What is missing now in terms of virtualization features?

Here’s what’s still missing or less mature on Apple Silicon today:

- Nested / Multi-Level Virtualization
- Full Hardware-Assisted I/O Virtualization (VT-d / AMD-Vi)
- Advanced Memory Management / Nested Page Tables
- OS/Hypervisor Ecosystem Integration

Apple Silicon virtualization is good for developer VMs, testing and macOS/Linux guests but it doesn’t match x86 hardware features for full enterprise or cloud-class virtualization especially when you need nested VMs, GPU passthrough or I/O virtualization.

 

[OptimusGrime]

Tech common in current x86 ecosystem that is missing or less developed on current M3 Ultra, M4 Max, M4 Pro, M5 and A19 Pro chips that may appear in future M chips:

- High-Precision AI Compute (FP8 / BF16 / Tensor Support)

M5 has Tensors and BF16 has been available since the M2 I believe.

- Dedicated AI/GPU Interconnects (NVLink / SXM / CXL)
- HBM (High-Bandwidth Memory)
- FPGA / Reconfigurable Hardware (common on servers)
- Hardware Virtualization Acceleration (VT-x, AMD-V) Parity

Parity for what hardware? If there are no dgpus what would they pass through? Nested virtualization is there in the hardware.

- Specialized Video & Encoding Blocks

What? Apple Silicon absolutely has specialized video encoding blocks.

 

[name99]

Tech common in current x86 ecosystem that is missing or less developed on current M3 Ultra, M4 Max, M4 Pro, M5 and A19 Pro chips that may appear in future M chips:

- High-Precision AI Compute (FP8 / BF16 / Tensor Support)
- Dedicated AI/GPU Interconnects (NVLink / SXM / CXL)
- HBM (High-Bandwidth Memory)
- FPGA / Reconfigurable Hardware (common on servers)
- Hardware Virtualization Acceleration (VT-x, AMD-V) Parity
- Specialized Video & Encoding Blocks
- Driver Ecosystem Maturity (especially for Open-source ML)
- Multi-Chip Scaling (Chiplets with High-Speed Interconnects)

But based on my informed expectations we can see these occuring

- Larger Neural Engines with wider precision support
- More GPU cores + bigger cache hierarchies
- Increased unified memory ceilings (2–4 TB)
- Better Metal/ML framework optimization
- Smarter heterogeneous scheduling

But fundamental changes like HBM, NVLink-style scaling or user-expandable memory are not likely because they conflict with Apple’s design philosophy of tight integration, efficiency and simplicity.

BF16 is definitely present on CPU and GPU. I'm not sure about ANE.
Tensor support (I assume by this you mean high perf matrix multiply) is present on all three.

FusionLink is the equivalent of NVLink

Apple gets the performance they need (bandwidth equivalent to HBM in the equivalent desktop HW) via many LPDDR5X channels. This *may* run out of steam - the main thing HBM gives you is overcoming the shoreline limitation of Apple's type of solution. But there are many alternative ways to continue growing via Apple's path.

You do realize that Apple runs their machines as hypervisors TODAY, right? The innermost most protected part of Darwin, the part that controls page tables, runs as a hypervisor, and has since at least A10. I'm unaware of any serious complaints about the performance of any third party virtualized OS (eg Linux, or a second copy of macOS) compared to x86.

I'm not going to bother with the rest. This feels like massive clickbait, or truly serious ignorance, of the sort that doesn't want to be cured.

 

[name99]

Apple Silicon will continue to add Neural Engine improvements, GPU cores, bigger UMA and better Metal/ML scheduling.

But enterprise/server features (ECC, multi-socket, NUMA control, redundant PSUs, ultra-fast networking) are by design outside Apple’s scope.

The trade-off is efficiency, simplicity and tight integration which is exactly why Apple doesn’t target HPC or data center workloads the same way x86 platforms do.

Umm, you do know that Apple are not just building servers, but data warehouses right now?
And have multiple patents describing what they are doing, patents that clearly build on the strengths of their SoCs, and do frequently do things rather differently than x86, though ultimately solving the same problems.

 

[crazy dave]

Safari.
View attachment 2590147

I have no idea! The changes all look like they're in presentation, not content, but since someone (github?) seems determined not to let me see them, who knows?

I took me several times (I got a unicorn saying it was taking too long to load), but eventually I got through. Not sure what's going on.

 

[Populus]

Hey! Welcome to 2026 Apple Silicon speculation thread!

Let’s reignite the speculation when we’re -supposedly- on the verge of Apple releasing the whole M5 Pro/Max/Ultra family, with the leaked hardware codes found on the Release Candidate of iOS 26.3

The hardware codes belong to the M5 Max and M5 Ultra SoC, which has fueled the speculation around which devices will be unveiled soon. There’s no trace of an M5 Pro.

Then, YouTuber Max Yuryev and their family’s channel proposed a new theory.

We’ve been hearing rumors of a separate block of GPU cores for the M5 pro and max chips, with a stacked 2.5G architecture, right? What if, in order to achieve better scalability, the M5 Pro is just an M5 Max with half the GPU cores, with the same footprint? That way, all chips would be technically M5 Max, but some of them would have half the GPU cores, thus, being marketed and sold as M5 Pro chip.

This is the theory that has been thrown by MaxTech. What do you think? I think this is more likely to be the case rather than… I don’t know, Apple releasing only the Mac Studio and not the M5 Pro MacBook Pro. Or even releasing a MacBook Pro with both an M5 Max and an M5 Ultra… not going to happen.

MacBook Pros are likely coming with an M5 Pro and M5 Max SoC, but both chips are going to share the same footprint and the same identifier.

Also, the presence of an M5 Ultra code means that we’re likely to also get new Mac Studios with M5 Max and M5 Ultra SoC… hopefully they will release the new M5 and M5 Pro Mac mini, along with the new Apple Studio Display.

By the way, now that I mention the Mac mini… here’s some food for thought: what if the next Mac mini is only released with the base M5 chip and not the M5 Pro? If the M5 Pro ends up being an M5 Max with half the GPU cores, maybe it’s thermal profile is too high for the small Mac mini form factor. Remember how hot the M4 Pro Mac mini runs when properly pushed…

 

[Luftkopf]

How hot actual is the M4 Pro Mac mini? Surely it can’t be as hot as my old 11” i7 MacBook Air, or my friend’s 2019 i9 MacBook Pro.

 

[diamond.g]

How hot actual is the M4 Pro Mac mini? Surely it can’t be as hot as my old 11” i7 MacBook Air, or my friend’s 2019 i9 MacBook Pro.

By hot, do you mean TDP, or chip temp?

 

[DrWojtek]

Very exciting times for sure. The gpu leap this gen will surpass everything last years and probably for the future, too. Desktop users (that wont benefit from OLED panels) should consider this one, I know I do.

 

[tenthousandthings]

We’ve been hearing rumors of a separate block of GPU cores for the M5 pro and max chips, with a stacked 2.5G architecture, right? What if, in order to achieve better scalability, the M5 Pro is just an M5 Max with half the GPU cores, with the same footprint? That way, all chips would be technically M5 Max, but some of them would have half the GPU cores, thus, being marketed and sold as M5 Pro chip.

This is the theory that has been thrown by MaxTech. What do you think? I think this is more likely to be the case rather than… I don’t know, Apple releasing only the Mac Studio and not the M5 Pro MacBook Pro. Or even releasing a MacBook Pro with both an M5 Max and an M5 Ultra… not going to happen.

MacBook Pros are likely coming with an M5 Pro and M5 Max SoC, but both chips are going to share the same footprint and the same identifier.

That sounds like the “chop” concept (Jade Chop, Rhodes Chop, Brava Chop). Same CPU, half the GPU. But "chop" was always a sort of misnomer, because the Max was really just a Pro with maximum memory bandwidth and maximum GPU. The Pro was the base for the Max, not the other way around.

I think he's right about there being a new, uniform "footprint" that comes with SoIC. TSMC even hints at something like that in their public description of SoIC: "From the outside, the integrated chip looks like an ordinary SoC. But on the inside, it is embedded with heterogeneously integrated functionalities."

I will be disappointed if Apple doesn't press forward with variations for more of the IC elements, not just the GPU.

 

[tenthousandthings]

Also, the presence of an M5 Ultra code means that we’re likely to also get new Mac Studios with M5 Max and M5 Ultra SoC… hopefully they will release the new M5 and M5 Pro Mac mini, along with the new Apple Studio Display.

By the way, now that I mention the Mac mini… here’s some food for thought: what if the next Mac mini is only released with the base M5 chip and not the M5 Pro? If the M5 Pro ends up being an M5 Max with half the GPU cores, maybe it’s thermal profile is too high for the small Mac mini form factor. Remember how hot the M4 Pro Mac mini runs when properly pushed…

The presence of M5 Ultra in the build might just be a way for Apple to indicate it exists (and M4 Ultra does not exist), but it doesn't necessarily mean it will launch with macOS 26.3.

On the Mini, both J873g and J873s model identifiers have leaked, so it seems unlikely only one of them would launch.

 

[Luftkopf]

By hot, do you mean TDP, or chip temp?

I’ve just seen multiple mentions of the M4 Pro Max Mini running hot, so I’m wondering what people are seeing? How do they compare to Intel Macs?

 

[Populus]

I’ve just seen multiple mentions of the M4 Pro Max Mini running hot, so I’m wondering what people are seeing? How do they compare to Intel Macs?

I only say it because I had both the M4 and the M4 Pro Mac mini, and the latter was a hot metal box, while the former had the typical slightly warm touch any Mac mini has had. The M4 Pro ran the fan more often and louder than the regular M4.

I even tried to put an aluminum heat dissipating grill on top of the M4 Pro Mac mini to try to lower it’s temperature because it seemed excessive.

I ended up returning both and waiting for the M5 model.

 

[Luftkopf]

I only say it because I had both the M4 and the M4 Pro Mac mini, and the latter was a hot metal box, while the former had the typical slightly warm touch any Mac mini has had. The M4 Pro ran the fan more often and louder than the regular M4.

I even tried to put an aluminum heat dissipating grill on top of the M4 Pro Mac mini to try to lower it’s temperature because it seemed excessive.

I ended up returning both and waiting for the M5 model.

I’m curious about the cooling capacity of the mini. I’ve always had to run custom fan curves to keep the Intel Macs cool, so I’m wondering whether it’s sufficient to run the fans in the Pro mini faster or if the M4 Pro just runs too hot for it.

You didn’t happen to keep an eye on the CPU/GPU temps did you?

 

[IvyKing]

How hot actual is the M4 Pro Mac mini? Surely it can’t be as hot as my old 11” i7 MacBook Air, or my friend’s 2019 i9 MacBook Pro.

I know from experience that the fans on an M4 Pro Mini do get loud when the CPU's are getting exercised. Concerns about cooling and the elimination of USB-A ports were two concerns I had about the smaller form factor. I was perfectly content with the size of my M1 Mini.

My recollection was that the case got noticeably warm to the touch, but not quite as hot as an Intel MacBook.

 

[Populus]

You didn’t happen to keep an eye on the CPU/GPU temps did you?

Nah, if I did, this was more than a year ago, and I don’t remember…

But for sure, in terms of performance per watt, Apple Silicon is miles ahead of what Intel was.

 

[splifingate]

Currently (and thoroughly) enjoying my M2 Max Studio, and patiently awaiting the extancy of M5 Max/Ultra in the Studio form-factor. My little beastie usually hovers around 40C (currently 39C after all-day, intensive usage).

If AAPL decides to continue to differentiate/fragment, I can easily wait for M6/7/8....

 

[Luftkopf]

Currently (and thoroughly) enjoying my M2 Max Studio, and patiently awaiting the extancy of M5 Max/Ultra in the Studio form-factor. My little beastie usually hovers around 40C (currently 39C after all-day, intensive usage).

If AAPL decides to continue to differentiate/fragment, I can easily wait for M6/7/8....

I'm currently dreaming of the rumoured M5 Max iMac. I'm not going to be able to afford one, and my M3 Pro is showing no signs of struggling, but I'll keep dreaming.

It's that, or an M3 Ultra, a refurbished 16" M3 Max MacBook Pro, and an M3 iPad Air, just to collect the whole set. Now that will never happen unless I win the lottery!

 

[Populus]

I won’t be able to afford an M5 Max or Ultra Mac Studio, but as you said, one can dream. And my mind cannot imagine the vastness of the raw power of such machine. Incredible…

 

[splifingate]

...and maybe the M5 Max will have the option of 512GB system RAM...wallets are gana get crushed!

 

[Populus]

Okay, we’ve seen that just as predicted on this thread, the M5 received a focus on the GPU, on its performance and the AI processing capabilities of the GPU.

Do you think… I know it’s too early to comment on this and it’s pure speculation, but do you think the M6 generation will be a small upgrade focused on just adopting the 2nm tech, with minor improvements to performance and efficiency? Maybe they will focus on the Neural Engine this time? Or will they use the jump to this new tech (including the new GAAFET transistors) to introduce deep architectural changes.

I personally think the Neural Engine will remain quite untouched, aside from the gains of a higher clock speed. But I see Apple more interested in enhancing the CPU and GPU neural accelerators for some reason. Also, the same way they implemented a completely new architecture for the M5 CPU e-cores, I think they will do the same with the M6 CPU p-cores, completing the revamp of the CPU architecture.

I’m not sure if the M6 update will be similar to the M3, which was a bit “in between” and short lived, or if the M5 will be that type of chip and the M6 will be a more solid foundation for the next generation Apple Silicon.