What can you run with 256 GB of memory?

A model does not get the 256 GB. It gets 192 GB on both machines here sold with 256 GB, because the system keeps the rest. That holds 38 of the 39 current models at 32k of context, and the strongest of them is GLM-5.3-Flash.

Run the numbers on GLM-5.3-Flash at 256 GB

Jump to: The machines sold with 256 GB · Models that fit in 256 GB · What 512 GB adds over 256 GB · How much context 256 GB leaves room for · Does a 256 GB machine pay for itself?

The machines sold with 256 GB

There are two machines here with 256 GB, and each hands a model the same amount. The fourth column is what each one holds at 32k of context, and it is the number that machine's own page prints.

List prices, and the memory each maker publishes. A machine marked previous is one that is no longer sold, priced at what it launched at.

Models that fit in 256 GB

Every current model a 256 GB machine here holds at 32k of context, strongest first. The third column is the whole job at once: the weights plus the cache for 32k of context. The fourth says how many of the machines at this size have room for it, which is the part a size on its own cannot tell you. Speeds are on the Mac Studio M5 Ultra, 256GB, the machine this list is cut from; speed follows memory bandwidth rather than memory size, so another machine at 256 GB runs the same model at its own rate, and its page gives every one of them.

For scale, the calculator starts from 80 tok/s for a hosted API and times a local machine against it. 22 of the 38 above reach it, and the slowest is 10 tok/s.

Superseded models are left out of the count, because what a machine is worth buying for is what you would run on it today; the leaderboard ranks every model this site lists, older ones included. Where the cache figure comes from is one section of the memory guide.

What 512 GB adds over 256 GB

The machine here with 512 GB that hands a model the most of it is the Mac Studio M5 Ultra, 512GB, at 384 GB, and it holds 39 of the 39. At 256 GB the most is 192 GB, on the Mac Studio M5 Ultra, 256GB, holding 38. The step adds Tencent Hy3. The strongest either way is GLM-5.3-Flash.

The cheapest machine at 512 GB is the Mac Studio M3 Ultra, 512GB at $9,499. The cheapest at 256 GB is the Mac Studio M5 Ultra, 256GB at $10,799, so the larger size is the cheaper of the two here.

Both sides are read off the machine at each size that hands a model the most, so the comparison is the best case against the best case. The table above has every machine at 256 GB and what each of them holds.

How much context 256 GB leaves room for

The cache grows with the window you ask for, so the same machine holds fewer models the longer the context. Every machine at this size hands a model 192 GB, so one column covers all of them. A model is counted only where its own context ceiling reaches that far.

Counted over the 39 current models, at the quantisation each is listed at and with the cache at 16 bits. The calculator has a switch for a smaller cache, which changes both what fits and how fast a long window runs.

Does a 256 GB machine pay for itself?

On the Mac Studio M5 Ultra, 256GB at $10,799, the cheapest machine at 256 GB with a published price, the model that pays it back soonest at 500k tokens a day is Inkling Small, in 124 years. At 20M tokens a day, agents running most of the day, it is Inkling Small in 3.1 years.

The sum is the same everywhere on this site: what the same work costs to rent, less what the electricity costs to generate it, against the price of the machine. The best buys rank the quickest pay-back at every level of use, and what it costs a month puts the machine and the API bill in the same shape.

Put your own usage in

Every figure is at 32k of context unless the row says otherwise, with the cache at 16 bits and each model at the quantisation this site lists it at. Weights are the published file sizes on each model's page, and the cache is worked out from the architecture recorded there. What a model gets is the memory the GPU can address, which each machine's page explains. Speeds say whether anybody measured them; where they were not, they are worked out from memory bandwidth. To change the context, the quantisation or the price you would pay, open the calculator. For the sizes either side of this one, the memory guide has the ladder in full.