llamaperf
Submit

M1 Max 32GB

APPLE · 32GB unified memory · 2 reports

On the M1 Max 32GB, Qwen3.8 27B at 4-bit, the model most run on it, generates a median of 15.8 tokens per second from one plain run (community reports, as of 29 Oct 2026).

As of 29 Oct 2026, the models most run on the M1 Max 32GB, with generation and prompt-processing medians from plain runs (one device, one request, no speculative decoding):

  • Qwen3.8 27B
    15.8gen t/s at 4-bit · one run
    PP 81.8 t/s · 1 run

Engines people use on it: MLX 1 (reports)

See reported performanceEstimate what fits
How to compare these reports →Filter reports by setup →All M1 Macs compared →

Run models on your M1 Max 32GB? Add your numbers.

Your coding agent can run a speed test on your machine and send it here with a free account. Or paste a result you've already posted.

Benchmark with your agentPaste a result

Reported performance on the M1 Max 32GB

These reports mix GPU counts and memory setups, so a model listed here may have needed extra cards or CPU offloading. The calculator estimates which models fit in 32 GB of unified memory.

This page is thin (2 of 3 reports needed for indexing). Help fill it in.
Date ↓Date ↑Token speed ↓Token speed ↑Prefill ↓Prefill ↑Context ↓Context ↑Ranked
Tone: positiveApple hardware
generation:
12-15 tokens/s
quant:
IQ2_K (MLX)
kv:
KVarN K4/V4

Reported by the source; GPU count, offloading and concurrent requests can change these figures. Prompt-processing speed is shown only when the source reports it; generation speed cannot tell us time to first token. Check the full setup before comparing.

codingmath

User reports Qwen3.8-Flash-Next (125B) at 12-15 decode tokens per second on a 2021 M1 Max with 32 GB of unified memory. Setup is MoEspresso 3 with IQ2_K routed projections and KVarN K4/V4 KV cache, using Metal execution through the mlx-iqk library. Cache-Prior expert bias is enabled by default; prefill is unbiased. The user notes the speed holds only when no other memory-hungry applications are running. The user also reports an overall score of 84.3% on a reproducible 48-question benchmark, ahead of Claude Opus 4.8 xhigh (80.3%) and other hosted solutions.

Sep 27, 2026
Apple hardware
generation:
15.8 tokens/s
prompt processing (prefill):
81.8 tokens/s
quant:
4bit

Reported by the source; GPU count, offloading and concurrent requests can change these figures. Prompt-processing speed is input throughput, not time to first token. Check the full setup before comparing.

User compares MLX and llama.cpp on Qwen3.8-27B. MLX reaches 15.81 t/s generation and 81.76 t/s prompt processing. llama.cpp reaches 9.69 t/s generation and 99.61 t/s prompt processing. The llama.cpp run uses the UD-Q4_K_M quant.

Sep 9, 2026

Get a weekly email of new M1 Max 32GB reports and newly released models that fit it.

Email me new reports

Reported configurations

Individual observations from this page, not expected speeds or a ranking. Select a model to open its report, with the source and offloading conditions. Reported context may be a limit rather than actual input length.

Model / hardwareQuant / engineReported contextGeneration
Qwen3.8 27B
M1 Max 32GB
4bit
MLX
Not reported15.8 tokens/s

Frequently asked

How fast is the M1 Max 32GB for local LLMs?

On the M1 Max 32GB, Qwen3.8 27B at 4-bit, the model most run on it, generates a median of 15.8 tokens per second from one plain run (community reports, as of 29 Oct 2026).