Can an Apple M4 Max 128GB run GPT OSS 120B?

In practice

GPT OSS 120B sits comfortably on Apple M4 Max 128GB. At MXFP4 it takes 59.65 GB of the 96 GB available, which leaves 36.35 GB spare, enough for the 8k window we assumed and for a window of up to about 128k tokens before memory runs out. Generation runs at about 68.3 tokens per second, well past the speed anyone reads at, so a 500-token answer takes about 7 seconds.

Of the memory in use, 99% is the weights and 0% is the attention cache. There is no higher-quality file to step up to, so the spare memory is best spent on a longer window.

Reading a 4,000-token prompt, roughly a long email or a few pages of text, takes about 8 seconds before the first word appears. That part is limited by compute and not by bandwidth, so it follows a different spec from the writing speed.

The numbers

GPT OSS 120B has 116.83 billion parameters across 36 layers, of which 5.7 billion are read for each token. The Apple M4 Max 128GB has 128 GB of unified memory at 546 GB/s, of which about 96 GB is left for a model once the operating system has taken its share.

1 of the 1 levels fit entirely. Levels marked degraded are listed for completeness, not as advice. This model is published in a single native format, so every row is the same file.

How long a conversation it holds

Memory rises with the length of the conversation because the attention cache keeps a key and a value for every token. At MXFP4, the longest window that still fits on this device is 128k tokens. Setting the window to what you need and no more is the cheapest way to buy back memory.

The settings we would use

llama-server -m <model>.gguf -ngl all -c 8192 -fa on

In Ollama, set OLLAMA_CONTEXT_LENGTH to the window you want, run ollama ps while the model is loaded and check that the processor column reads 100% GPU. If it shows a split on a model that fits, the Ollama guide shows how to fix it. The llama.cpp flags guide explains each option above.

What else an Apple M4 Max 128GB runs well

For comparison, these are the other popular models that fit this device at a sound quantization, best pairings first.

The same model on nearby hardware

Memory decides whether GPT OSS 120B loads and bandwidth decides how quickly it answers. Here is what a step down, a card with the same memory and a step up do with it.

Wait for the first word1.1 s

Then writes at68.3 tok/s

Whole answer3.6 s

YouWhy does my model use more memory when the conversation gets longer?

Model

Simulated from our estimate at a 512-token question, not a recording. Assumes nothing else is competing for the GPU.

Questions people ask

Keep going

Everything the Apple M4 Max 128GB runs, the full breakdown for GPT OSS 120B, how the numbers are made in the methodology, or a different pairing in the calculator.