Apple M1 Max for local LLMs
2 measured results from 2 builds, each linked to its source.
| Memory | unified memory |
|---|---|
| Memory bandwidth | not listed |
| Typical EU price | not tracked yet |
Can it run? gpt-oss 20B on Apple M1 Max
All builds with the Apple M1 Max
| Build | GPU(s) | Memory GB (total) | Results |
|---|---|---|---|
| MacBook Pro M1 Max 32-core GPU, 64 GB | Apple M1 Max 32-core GPU | 64 | 1 |
| Apple M1 Max 64GB | Apple M1 Max (unified) | 64 | 1 |
All measured results
(MTP, DFlash, DSpark and similar draft-and-verify methods give much higher tok/s on predictable text and are not comparable 1:1 with plain decode)
| Model | Build | Ran on | Quant | Engine | tok/s | Details |
|---|---|---|---|---|---|---|
| Llama 3 8B | MacBook Pro M1 Max 32-core GPU, 64 GB | Apple M1 Max | Q4_K_M | llama.cpp | 34.49 | |
| gpt-oss 20B | Apple M1 Max 64GB | Apple M1 Max | MXFP4 | llama.cpp | 75.15 |
All Apple M1 Max builds: see also all GPUs · best LLMs by memory size · download the dataset.
FAQ
How fast is the Apple M1 Max for local LLMs?
The fastest plain-decode result we have is 75.15 tok/s on gpt-oss 20B (MXFP4, llama.cpp) in Apple M1 Max 64GB. Speed depends heavily on the model, quant and engine, so compare like with like in the table. RigBase lists 2 measured results from 2 builds.
How much memory does the Apple M1 Max have?
unified memory.
Which models have been measured on the Apple M1 Max?
Llama 3 8B, gpt-oss 20B.
Where do these numbers come from?
Every row links to a build page with its original public source (benchmark logs, READMEs, forum posts). RigBase does not re-run the benchmarks. Results marked ⚡ use speculative decoding and are not comparable 1:1 with plain decode.
Referral link: if you sign up with Google and add $10, Runpod gives you and RigBase bonus credit (terms). It never affects our data.