upvote
(I'm not one of the people you're speaking of with a 128gb M5 but) if you want to run one of the medium-sized open-weights models (Qwen 27b, 35b, Gemma 4 26b, 31b) or larger, you get into an interesting optimisation space.

* yes, you can run it on an older/smaller GPU plus system RAM but performance will suffer

* if you want optimal GPU performance you need the model in VRAM plus context, so 24GB (3090, 4090) or 32GB (5090) cards, plus a system that's reasonable powerful to plug them in to. Ideally you'd have a multiple cards working together but for optimal performance this means either 2x 3090 or nvidia's workstation cards.

* you can go for a 128gb Strix Halo system, but the memory bandwidth isn't great and they're becoming increasingly more expensive (5.5k EUR for HP laptop, 3.9k EUR for GMKtec EVO-X2 mini PC)

* you can go for a 128gb DGX Spark (5k EUR+) which also has unspectacular memory bandwidth or RTX Spark (price unclear but probably not cheaper)

* or go for a Mac with a decent CPU and a good amount of RAM (bandwidth varies by model, but typically a bit better than Strix Halo/DGX Spark and worse than bespoke GPUs.

As usual with such questions, there are of course cheaper paths (if you want to accept the tradeoffs) but Macs are reasonable vs. competition for these workloads.

reply
I just recently got into experimenting with local LLMs when I had anyway (for non-LLM reasons) built myself a new desktop system with Intel Ultra 270K-Plus and RTX 5080. With 64GB system RAM and 16GB VRAM. Relatively speaking a high-performing and low-to-moderate cost system.

I wasn't really expecting much from these local open weight models neither when it comes to speed or "intelligence", but my preconceptions were quickly put ashame when I got ollama up and running and pulled my first model. I get a consistent 117-128 t/s with Gemma4:26b-a4b without any tuning (just the default settings), which was much faster than I had expected. Can't wait to dive deeper into this, especially with Qwen3.6 models.

Does anyone's have experience adding a 2nd Nvidia GPU of the same generation but different (slower) model in the same system? Will it give a major boost with larger models, or will the slower card just be a bottleneck? I have an unused RTX 5060 Ti 16GB that I'm considering to install alongside the RTX 5080, but it would necessitate removing some other hardware, so I haven't bothered yet.

reply
I'd say adding another 16Gb gpu would be worth it - you'd be able to run larger model/larger context all within gpu's. It would give you more options of what you can run fast. Your current model probably doesn't run completely from GPU (depending on quants I don't think you can squeeze Gemma4:26b into 16Gb vram), so you already have some layers running on gpu and some on cpu. If you add another gpu you might be able to move all layers to vram which should speed up things for you. The layers calculations happen on whatever gpu's it sits, so the layers that are already on your rtx5080 would compute same, but the layers that currently your cpu handles will be computed with faster vram/compute of rtx5060.
reply
And with a mac, there are no cuda drivers to fiddle with.
reply
But prompt processing is terrible
reply
A mac with a boatload of RAM can run models that will exceed the limits of any GPU not worth at least twice the Apple hardware itself.

You get fewer tokens per second, but at some point the balance between quality and quantity makes the large model size worth the spend.

When you're spending this kind of money, you may as well treat yourself to a pretty screen and some decent speakers. Nothing the competition doesn't offer these days, but you get them for free with the car-priced RAM upgrade so why go for less.

reply
I don't even travel a ton but portability is huge. It's not a flex, it's a functional thing that lets me move around within my house or work while I'm at my parents or traveling or anywhere else. Other than my media collection that lives on my home server, I want most of my files to come with me on my laptop.
reply
The fact that I can take it with me? That I don’t need internet to still have access to deepseek? The fact that electricity is expensive and an mbp uses ~10% of the power that an equivalent vram set up would using gpu’s. Also, in order to get the same vram I would need to spend a similar amount, but wouldn’t also have a machine that was useful for other workloads that need a huge amount of ram.
reply
Potentially going to sound privileged here, but why not both?

Personally when going on the road I like portability (14" MBP or MBA), but at home I want raw non-thermally throttled power.

reply
I think it is because desktop computers with GPUs with enough VRAM to run interesting models are insanely expensive, hard to source and consume a lot of electricity and dissipate a lot of heat.
reply
What GPU can I buy with >100GB of memory?
reply
DGX Spark is one, but really depends on how much you want to spend
reply
273GB/s bandwidth vs 614 GB/s of the M5 Max. And you're getting a whole laptop.

$5k for DGX Spark as well.

reply
Prompt processing time is better on the spark, which aligns more with coding (more reading than writing).

I spent less than $4k, OEM are better boxes for cooling, no apple markup, I get a real Linux system for stuff like k3s.

reply
Yes, it's better on the Spark but the M5 is a lot closer than before with neural accelrators. After prompt processing, token generation speed on the M5 Max is 2.3x faster.

No Apple markup but you get the Nvidia market up instead. Prior to the recent Apple price increase due to RAM shortage, an M5 Max 128GB was a bargain if you want to run local LLMs.

reply
I can get 2.5 spark for the price of the M5, will have better throughput and access to bigger models (more vram when running tensor parallel)
reply
I have a bunch of computers and gadgets, why settle on one?
reply
Unified memory.
reply
Yeah, it's a much better idea to buy many used 3090s. 4090s or 5090s if you can afford it. Way faster.
reply
Probably depends on what you're trying to do.

You need an expensive motherboard, cooling, PSU(s) to use multiple high end GPUs together. Then there is the noise and the fact that you can't bring it on an airplane.

reply