llamaperf

What people run, by hardware

Pick the kind of machine you have. You'll see the model sizes people report running on it, most reported first, with the speed they typically get. It tells you what works on hardware like yours. It doesn't tell you which model gives the best answers.

To check a specific model on your exact card, use the calculator. For coding, see the coding models by VRAM.

Maker
Mac

A Mac counts when its report names the product, or when only that product was sold with its chip and memory.

Apple Silicon
Model size

Apple Silicon 18-36 GB· Mac Studio only

Mid-size Macs. Up to about 30B-class models at 4 bits. Memory speed 410 to 461 GB/s across the 2 devices with reports, so a typical figure here spans all of them. 12 single-device reports.

#ModelReports
1Qwen3.827B
Alibaba · mostly 4-bit
Gen 21 t/s (2 runs) · PP 82 t/s (1 run) · with speculation 38 t/s (1)
5
2Qwen3.59B
Alibaba · mostly 4-bit
Gen 80 t/s (1 run) · with speculation 116 t/s (1)
4
3Qwen3.635B · 3B active
Alibaba · mostly 4-bit
No generation data · with speculation 173 t/s (1)
3

How this list works

Each row is one model size, because a 27B and a 397B from the same family are different choices. Rows are ordered by how many people report running that size on this hardware.

Generation speed is the median of runs on one device, serving one request, without speculative decoding. PP is the median prompt-processing speed from plain runs that reported it, so its run count can be smaller. Speculative decoding (MTP, a draft model, DFlash, n-gram lookup) can make the same model several times faster, so those runs get their own column. A report that doesn't make clear which kind of run it was counts in neither.

Hardware is grouped by how much it holds and how fast its memory is, since memory speed sets how fast a model writes. Click a model for every report behind its row.