← All MacBook Pro 16" models

MacBook Pro 16" M4 Max

The M4 Max MacBook Pro 16" runs local models at 410 to 546 GB/s of memory bandwidth with 36 to 128 GB of unified memory. On Apple Silicon that memory is shared with the GPU, so the whole pool is available for weights: at 128 GB you can hold roughly a 172B dense model at Q4. The Max is where the bus gets wide enough that model size, not bandwidth, becomes the thing you plan around.

Apple no longer sells this configuration new. It stays fully evaluated here because the used market is where most of its local AI value now sits.

Specifications

ChipApple M4 Max
CPU cores14 or 16
GPU cores32 or 40
Unified memory36, 48, 64, or 128 GB
Memory bandwidth410 to 546 GB/s
Neural Engine38 TOPS
Released2024
AvailabilityUsed market

Memory bandwidth is faster than 72% of the Apple Silicon chips shipped in a Mac, against a 819 GB/s peak.

Its memory ceiling is above 67% of them, against a 512 GB peak.

Pick your configuration

Every option Apple sells with this chip. The model list below recomputes against the one you pick.

GPU cores

Unified memory

Apple couples memory to the core count on this chip, so the options change with the bin above.

What each memory option runs

Unified memory is the ceiling and it is soldered, so this is the decision you cannot revisit.

36 GB unified memory

31 GB usable for weights · 410 GB/s

3,417 of 3,641 models fit, and 3,252 of them run with headroom rather than as a squeeze.

Largest model at Q4
Chinese Mixtral 8x7B · 46.91B
Best all-round pick
Qwen3.5 0.8B · Q8_0 · ~247 tok/s

48 GB unified memory

42 GB usable for weights · 546 GB/s

3,503 of 3,641 models fit, and 3,351 of them run with headroom rather than as a squeeze.

Largest model at Q4
Yi 34Bx2 MoE 60B DPO · 60.81B
Best all-round pick
Qwen3.5 4B · Q8_0 · ~61 tok/s

64 GB unified memory

56 GB usable for weights · 546 GB/s

3,541 of 3,641 models fit, and 3,385 of them run with headroom rather than as a squeeze.

Largest model at Q4
Kimi K2.6 JANGTQ_3L · 83.44B
Best all-round pick
Qwen3.5 4B · Q8_0 · ~61 tok/s

128 GB unified memory

112 GB usable for weights · 546 GB/s

3,582 of 3,641 models fit, and 3,532 of them run with headroom rather than as a squeeze.

Largest model at Q4
Kimi K2.5 · 171B
Best all-round pick
Qwen3.6 35B A3B · Q8_0 · ~95 tok/s

What a 36 GB M4 Max MacBook Pro 16" can run

Every model in the database against this exact configuration, at 410 GB/s. Ratings and speeds are the same numbers the model pages show.

Showing 3641 of 3641 models

Multimodal · Alibaba · 2026-02-28

Q8_0Excellent
1.5 GB4% of RAM~247 tok/sEstimated0.87B params
Run with ToolPiper

Multimodal · Alibaba · 2026-02-28

Q8_0Excellent
3.0 GB8% of RAM~95 tok/sEstimated2.27B params
Run with ToolPiper

Multimodal · Alibaba · 2026-02-28

Q8_0Excellent
1.5 GB4% of RAM~247 tok/sEstimated0.87B params
Run with ToolPiper

Multimodal · Alibaba · 2026-02-28

Q8_0Excellent
3.0 GB8% of RAM~95 tok/sEstimated2.27B params
Run with ToolPiper

Multimodal · Alibaba · 2026-02-27

Q8_0Excellent
5.7 GB16% of RAM~46 tok/sEstimated4.66B params
Run with ToolPiper

Multimodal · Alibaba · 2026-02-27

Q8_0Excellent
5.7 GB16% of RAM~46 tok/sEstimated4.66B params
Run with ToolPiper

General · Liquid AI · 2025-11-28

Q8_0Excellent
1.8 GB5% of RAM~184 tok/sEstimated1.17B params
Run with ToolPiper

General · Liquid AI · 2025-11-28

Q8_0Excellent
0.9 GB2% of RAM~614 tok/sEstimated0.35B params
Run with ToolPiper

Reasoning · Liquid AI · 2025-11-28

Q8_0Excellent
1.8 GB5% of RAM~184 tok/sEstimated1.17B params
Run with ToolPiper

General · Liquid AI · 2025-11-28

Q8_0Excellent
9.8 GB27% of RAMBenchmark needed8.3B params
Run with ToolPiper

General · Liquid AI · 2025-11-28

Q8_0Excellent
1.8 GB5% of RAM~184 tok/sEstimated1.17B params
Run with ToolPiper

General · Liquid AI · 2025-11-28

Q8_0Excellent
0.9 GB2% of RAM~614 tok/sEstimated0.35B params
Run with ToolPiper

General · Liquid AI · 2025-11-28

Q8_0Excellent
3.4 GB9% of RAM~84 tok/sEstimated2.57B params
Run with ToolPiper

General · Liquid AI · 2025-11-28

Q8_0Excellent
1.3 GB4% of RAM~290 tok/sEstimated0.74B params
Run with ToolPiper

General · Liquid AI · 2025-11-28

Q8_0Excellent
1.8 GB5% of RAM~184 tok/sEstimated1.17B params
Run with ToolPiper

General · Liquid AI · 2025-11-28

Q8_0Excellent
0.9 GB2% of RAM~614 tok/sEstimated0.35B params
Run with ToolPiper

General · Liquid AI · 2025-11-28

Q8_0Excellent
3.4 GB9% of RAM~84 tok/sEstimated2.57B params
Run with ToolPiper

General · Liquid AI · 2025-11-28

Q8_0Excellent
0.9 GB2% of RAM~614 tok/sEstimated0.35B params
Run with ToolPiper

General · Liquid AI · 2025-11-28

Q8_0Excellent
1.8 GB5% of RAM~184 tok/sEstimated1.17B params
Run with ToolPiper

General · Liquid AI · 2025-11-28

Q8_0Excellent
0.9 GB2% of RAM~614 tok/sEstimated0.35B params
Run with ToolPiper

General · Liquid AI · 2025-11-28

Q8_0Excellent
1.8 GB5% of RAM~184 tok/sEstimated1.17B params
Run with ToolPiper

Reasoning · Liquid AI · 2025-11-28

Q8_0Excellent
0.9 GB2% of RAM~614 tok/sEstimated0.35B params
Run with ToolPiper

General · ibm-granite · 2025-09-16

Q8_0Excellent
8.2 GB23% of RAMBenchmark needed6.94B params
Run with ToolPiper

Chat · Liquid AI · 2025-11-28

Q8_0Excellent
1.8 GB5% of RAM~184 tok/sEstimated1.17B params
Run with ToolPiper

Chat · Liquid AI · 2025-11-28

Q8_0Excellent
3.4 GB9% of RAM~84 tok/sEstimated2.57B params
Run with ToolPiper

General · ibm-granite · 2025-09-16

Q8_0Excellent
4.1 GB11% of RAM~67 tok/sEstimated3.19B params
Run with ToolPiper

Reasoning · jackrong · 2026-03-16

Q8_0Excellent
11.3 GB31% of RAM~22 tok/sEstimated9.65B params
Run with ToolPiper

Chat · Liquid AI · 2025-11-28

Q8_0Excellent
1.8 GB5% of RAM~184 tok/sEstimated1.17B params
Run with ToolPiper

General · NCAI · 2025-12-29

Q8_0Excellent
8.6 GB24% of RAMBenchmark needed7.25B params
Run with ToolPiper

Reasoning · HuggingFace · 2025-07-08

Q8_0Excellent
3.8 GB11% of RAM~72 tok/sEstimated3B params
Run with ToolPiper

General · Liquid AI · 2025-11-28

Q8_0Excellent
2.2 GB6% of RAM~143 tok/sEstimated1.5B params
Run with ToolPiper

General · Liquid AI · 2025-11-28

Q8_0Excellent
2.2 GB6% of RAM~143 tok/sEstimated1.5B params
Run with ToolPiper

Multimodal · NCAI · 2025-12-29

Q8_0Excellent
9.0 GB25% of RAMBenchmark needed7.58B params
Run with ToolPiper

Multimodal · Alibaba · 2026-02-27

Q8_0Excellent
11.3 GB31% of RAM~22 tok/sEstimated9.65B params
Run with ToolPiper

Multimodal · Google · 2025-07-30

Q8_0Excellent
6.2 GB17% of RAM~42 tok/sEstimated5.1B params
Run with ToolPiper

Multimodal · Google · 2025-06-25

Q8_0Excellent
5.0 GB14% of RAM~54 tok/sEstimated4B params
Run with ToolPiper

Multimodal · Alibaba · 2026-02-26

Q8_0Excellent
11.3 GB31% of RAM~22 tok/sEstimated9.65B params
Run with ToolPiper

Multimodal · Liquid AI · 2025-11-28

Q8_0Excellent
1.0 GB3% of RAM~477 tok/sEstimated0.45B params
Run with ToolPiper

Multimodal · Liquid AI · 2025-11-28

Q8_0Excellent
2.3 GB6% of RAM~134 tok/sEstimated1.6B params
Run with ToolPiper

Multimodal · Liquid AI · 2025-11-28

Q8_0Excellent
3.8 GB11% of RAM~72 tok/sEstimated3B params
Run with ToolPiper

Multimodal · Liquid AI · 2025-11-28

Q8_0Excellent
2.3 GB6% of RAM~136 tok/sEstimated1.58B params
Run with ToolPiper

General · lgai-exaone · 2025-03-12

Q8_0Excellent
3.2 GB9% of RAM~89 tok/sEstimated2.41B params
Run with ToolPiper

Multimodal · google · 2026-05

Q8_0Excellent
13.8 GB38% of RAM~18 tok/sEstimated11.96B params
Run with ToolPiper

General · LG AI · 2025-07-15

Q8_0Excellent
1.8 GB5% of RAM~179 tok/sEstimated1.2B params
Run with ToolPiper

General · raidium · 2026-06-15

Q8_0Excellent
0.5 GB1% of RAM~10,738 tok/sEstimated0.02B params
Run with ToolPiper

General · NCAI · 2025-12-29

Q8_0Excellent
22.4 GB62% of RAMBenchmark needed19.6B params
Run with ToolPiper

Embedding · taide · 2026-06-12

Q8_0Excellent
0.8 GB2% of RAM~716 tok/sEstimated0.3B params
Run with ToolPiper

General · Alibaba · 2025-04-27

Q8_0Excellent
1.3 GB4% of RAM~286 tok/sEstimated0.75B params
Run with ToolPiper

General · Alibaba · 2025-04-27

Q8_0Excellent
5.0 GB14% of RAM~53 tok/sEstimated4.02B params
Run with ToolPiper

Multimodal · Google · 2025-07-30

Q8_0Excellent
9.4 GB26% of RAM~27 tok/sEstimated8B params
Run with ToolPiper

General · Alibaba · 2025-04-27

Q8_0Excellent
2.8 GB8% of RAM~106 tok/sEstimated2.03B params
Run with ToolPiper

Multimodal · zai-org

Q8_0Excellent
2.0 GB6% of RAM~161 tok/sEstimated1.33B params
Run with ToolPiper

Multimodal · Alibaba

Q8_0Excellent
2.9 GB8% of RAM~101 tok/sEstimated2.13B params
Run with ToolPiper

Multimodal · Microsoft

Q8_0Excellent
5.1 GB14% of RAM~52 tok/sEstimated4.15B params
Run with ToolPiper

General · Alibaba

Q8_0Excellent
2.2 GB6% of RAM~139 tok/sEstimated1.54B params
Run with ToolPiper

General · Alibaba

Q8_0Excellent
1.0 GB3% of RAM~438 tok/sEstimated0.49B params
Run with ToolPiper

Reasoning · NVIDIA · 2025-06-01

Q8_0Excellent
10.5 GB29% of RAM~24 tok/sEstimated9B params
Run with ToolPiper

Multimodal · openbmb

Q8_0Excellent
2.0 GB5% of RAM~165 tok/sEstimated1.3B params
Run with ToolPiper

Reasoning · DeepSeek

Q8_0Excellent
2.5 GB7% of RAM~121 tok/sEstimated1.78B params
Run with ToolPiper

Multimodal · rednote-hilab

Q8_0Excellent
3.9 GB11% of RAM~71 tok/sEstimated3.04B params
Run with ToolPiper

General · DeepSeek

Q8_0Excellent
18.0 GB50% of RAMBenchmark needed15.71B params
Run with ToolPiper

Multimodal · datalab-to

Q8_0Excellent
1.3 GB4% of RAM~311 tok/sEstimated0.69B params
Run with ToolPiper

General · ibm-granite

Q8_0Excellent
4.3 GB12% of RAM~63 tok/sEstimated3.4B params
Run with ToolPiper

Multimodal · moonshotai

Q8_0Excellent
18.8 GB52% of RAMBenchmark needed16.41B params
Run with ToolPiper

General · openbmb

Q8_0Excellent
1.7 GB5% of RAM~199 tok/sEstimated1.08B params
Run with ToolPiper

Multimodal · rednote-hilab

Q8_0Excellent
3.9 GB11% of RAM~71 tok/sEstimated3.04B params
Run with ToolPiper

General · hmellor

Q8_0Excellent
1.9 GB5% of RAM~173 tok/sEstimated1.24B params
Run with ToolPiper

General · distil-labs

Q8_0Excellent
0.9 GB2% of RAM~614 tok/sEstimated0.35B params
Run with ToolPiper

Multimodal · nanonets

Q8_0Excellent
4.7 GB13% of RAM~57 tok/sEstimated3.75B params
Run with ToolPiper

General · farbodtavakkoli

Q8_0Excellent
1.8 GB5% of RAM~177 tok/sEstimated1.21B params
Run with ToolPiper

General · Liquid AI

Q8_0Excellent
9.9 GB28% of RAMBenchmark needed8.47B params
Run with ToolPiper

General · nanonets

Q8_0Excellent
4.7 GB13% of RAM~57 tok/sEstimated3.75B params
Run with ToolPiper

General · nanbeige

Q8_0Excellent
4.9 GB14% of RAM~55 tok/sEstimated3.93B params
Run with ToolPiper

Multimodal · ibm-granite

Q8_0Excellent
5.0 GB14% of RAM~54 tok/sEstimated4B params
Run with ToolPiper

General · ibm-granite

Q8_0Excellent
7.9 GB22% of RAMBenchmark needed6.67B params
Run with ToolPiper
Q8_0Excellent
0.5 GB1% of RAM~21,476 tok/sEstimated0.01B params
Run with ToolPiper

Multimodal · moonshotai

Q8_0Excellent
18.8 GB52% of RAMBenchmark needed16.41B params
Run with ToolPiper

Multimodal · lkhl

Q8_0Excellent
2.7 GB7% of RAM~110 tok/sEstimated1.96B params
Run with ToolPiper

Multimodal · typhoon-ai

Q8_0Excellent
4.7 GB13% of RAM~57 tok/sEstimated3.75B params
Run with ToolPiper

General · jinaai

Q8_0Excellent
2.2 GB6% of RAM~139 tok/sEstimated1.54B params
Run with ToolPiper

General · Alibaba

Q8_0Excellent
2.9 GB8% of RAM~101 tok/sEstimated2.13B params
Run with ToolPiper

General · kristaller486

Q8_0Excellent
3.9 GB11% of RAM~71 tok/sEstimated3.04B params
Run with ToolPiper

Reasoning · typhoon-ai

Q8_0Excellent
2.9 GB8% of RAM~101 tok/sEstimated2.13B params
Run with ToolPiper

General · Upstage

Q8_0Excellent
9.5 GB26% of RAMBenchmark needed8.05B params
Run with ToolPiper

General · zstanjj

Q8_0Excellent
4.8 GB13% of RAM~56 tok/sEstimated3.82B params
Run with ToolPiper

General · ibm-granite

Q8_0Excellent
4.3 GB12% of RAM~63 tok/sEstimated3.4B params
Run with ToolPiper

General · nvidia

Q8_0Excellent
4.8 GB13% of RAM~56 tok/sEstimated3.83B params
Run with ToolPiper

General · weiboai

Q8_0Excellent
3.9 GB11% of RAM~70 tok/sEstimated3.09B params
Run with ToolPiper

General · x-izhang

Q8_0Excellent
4.1 GB11% of RAM~66 tok/sEstimated3.25B params
Run with ToolPiper

General · openai

Q8_0Excellent
3.1 GB9% of RAMBenchmark needed2.37B params
Run with ToolPiper

General · paddlepaddle

Q8_0Excellent
1.6 GB4% of RAM~224 tok/sEstimated0.96B params
Run with ToolPiper

General · twinkle-ai

Q8_0Excellent
4.8 GB13% of RAM~55 tok/sEstimated3.88B params
Run with ToolPiper

General · Liquid AI

Q8_0Excellent
0.9 GB2% of RAM~614 tok/sEstimated0.35B params
Run with ToolPiper

General · pfnet

Q8_0Excellent
1.9 GB5% of RAM~166 tok/sEstimated1.29B params
Run with ToolPiper

General · openbmb

Q8_0Excellent
1.7 GB5% of RAM~199 tok/sEstimated1.08B params
Run with ToolPiper

General · ibm-granite

Q8_0Excellent
4.3 GB12% of RAM~63 tok/sEstimated3.4B params
Run with ToolPiper

General · baidu

Q8_0Excellent
0.9 GB3% of RAM~597 tok/sEstimated0.36B params
Run with ToolPiper

General · jetbrains

Q8_0Excellent
14.1 GB39% of RAMBenchmark needed12.15B params
Run with ToolPiper

General · amd

Q8_0Excellent
2.2 GB6% of RAM~143 tok/sEstimated1.5B params
Run with ToolPiper

General · bytedance-seed

Q8_0Excellent
11.0 GB30% of RAMBenchmark needed9.37B params
Run with ToolPiper

General · arcee-ai

Q8_0Excellent
7.3 GB20% of RAMBenchmark needed6.12B params
Run with ToolPiper

General · adamlucek

Q8_0Excellent
1.9 GB5% of RAM~173 tok/sEstimated1.24B params
Run with ToolPiper

Coding · shahriarferdoush

Q8_0Excellent
1.9 GB5% of RAM~173 tok/sEstimated1.24B params
Run with ToolPiper

General · ahczhg

Q8_0Excellent
1.9 GB5% of RAM~173 tok/sEstimated1.24B params
Run with ToolPiper

General · abaryan

Q8_0Excellent
1.9 GB5% of RAM~173 tok/sEstimated1.24B params
Run with ToolPiper

General · etherll

Q8_0Excellent
1.3 GB4% of RAM~290 tok/sEstimated0.74B params
Run with ToolPiper

General · stanford-oval

Q8_0Excellent
4.7 GB13% of RAM~57 tok/sEstimated3.75B params
Run with ToolPiper

General · jetbrains

Q8_0Excellent
14.1 GB39% of RAMBenchmark needed12.15B params
Run with ToolPiper

General · flowaicom

Q8_0Excellent
4.8 GB13% of RAM~56 tok/sEstimated3.82B params
Run with ToolPiper

General · bllossom

Q8_0Excellent
4.1 GB11% of RAM~67 tok/sEstimated3.21B params
Run with ToolPiper

General · farbodtavakkoli

Q8_0Excellent
2.5 GB7% of RAMBenchmark needed1.77B params
Run with ToolPiper

General · z-lab

Q8_0Excellent
4.4 GB12% of RAM~62 tok/sEstimated3.48B params
Run with ToolPiper
Q8_0Excellent
4.0 GB11% of RAM~69 tok/sEstimated3.13B params
Run with ToolPiper

General · kamilamila

Q8_0Excellent
1.2 GB3% of RAM~346 tok/sEstimated0.62B params
Run with ToolPiper

General · radheneev

Q8_0Excellent
4.1 GB11% of RAM~67 tok/sEstimated3.21B params
Run with ToolPiper
Q8_0Excellent
2.2 GB6% of RAM~143 tok/sEstimated1.5B params
Run with ToolPiper

General · paddlepaddle

Q8_0Excellent
1.6 GB4% of RAM~224 tok/sEstimated0.96B params
Run with ToolPiper

General · moonshotai

Q8_0Excellent
18.8 GB52% of RAMBenchmark needed16.41B params
Run with ToolPiper

General · jangq-ai

Q8_0Excellent
17.6 GB49% of RAMBenchmark needed15.3B params
Run with ToolPiper
Q8_0Excellent
5.7 GB16% of RAM~46 tok/sEstimated4.66B params
Run with ToolPiper

General · agentica-org

Q8_0Excellent
2.5 GB7% of RAM~121 tok/sEstimated1.78B params
Run with ToolPiper

General · NousResearch

Q8_0Excellent
4.1 GB11% of RAM~67 tok/sEstimated3.21B params
Run with ToolPiper

Reasoning · nvidia

Q8_0Excellent
4.7 GB13% of RAM~57 tok/sEstimated3.75B params
Run with ToolPiper

General · inference-net

Q8_0Excellent
4.1 GB11% of RAM~67 tok/sEstimated3.21B params
Run with ToolPiper

General · novaciano

Q8_0Excellent
2.2 GB6% of RAM~143 tok/sEstimated1.5B params
Run with ToolPiper

General · dmusingu

Q8_0Excellent
2.9 GB8% of RAM~101 tok/sEstimated2.13B params
Run with ToolPiper
Q8_0Excellent
5.3 GB15% of RAM~50 tok/sEstimated4.3B params
Run with ToolPiper

General · kgrabko

Q8_0Excellent
2.2 GB6% of RAM~143 tok/sEstimated1.5B params
Run with ToolPiper

General · ordenwills

Q8_0Excellent
0.9 GB2% of RAM~614 tok/sEstimated0.35B params
Run with ToolPiper
Q8_0Excellent
2.1 GB6% of RAM~155 tok/sEstimated1.39B params
Run with ToolPiper

General · carsenk

Q8_0Excellent
1.9 GB5% of RAM~173 tok/sEstimated1.24B params
Run with ToolPiper

General · dealignai

Q8_0Excellent
17.6 GB49% of RAMBenchmark needed15.3B params
Run with ToolPiper

General · ibm-granite

Q8_0Excellent
3.8 GB11% of RAM~72 tok/sEstimated2.98B params
Run with ToolPiper

Coding · ibm-granite

Q8_0Excellent
4.4 GB12% of RAM~62 tok/sEstimated3.48B params
Run with ToolPiper

Reasoning · iffyuan

Q8_0Excellent
5.0 GB14% of RAM~53 tok/sEstimated4.07B params
Run with ToolPiper

Reasoning · nvidia

Q8_0Excellent
2.2 GB6% of RAM~139 tok/sEstimated1.54B params
Run with ToolPiper

General · zero-point-ai

Q8_0Excellent
3.0 GB8% of RAM~95 tok/sEstimated2.27B params
Run with ToolPiper

General · stanfordaimi

Q8_0Excellent
4.7 GB13% of RAM~57 tok/sEstimated3.75B params
Run with ToolPiper

General · trillionlabs

Q8_0Excellent
18.6 GB52% of RAMBenchmark needed16.24B params
Run with ToolPiper

General · Microsoft

Q8_0Excellent
4.8 GB13% of RAM~56 tok/sEstimated3.82B params
Run with ToolPiper
Q8_0Excellent
14.5 GB40% of RAMBenchmark needed12.59B params
Run with ToolPiper

General · huihui-ai

Q8_0Excellent
4.3 GB12% of RAM~63 tok/sEstimated3.4B params
Run with ToolPiper

General · ibm-granite

Q8_0Excellent
2.0 GB6% of RAMBenchmark needed1.33B params
Run with ToolPiper

General · openbmb

Q8_0Excellent
2.0 GB5% of RAM~165 tok/sEstimated1.3B params
Run with ToolPiper

General · ibm-granite

Q8_0Excellent
2.1 GB6% of RAM~147 tok/sEstimated1.46B params
Run with ToolPiper

General · osaurusai

Q8_0Excellent
11.3 GB31% of RAMBenchmark needed9.65B params
Run with ToolPiper

General · pyoakum

Q8_0Excellent
6.5 GB18% of RAMBenchmark needed5.35B params
Run with ToolPiper

General · ibm-granite

Q8_0Excellent
4.2 GB12% of RAMBenchmark needed3.3B params
Run with ToolPiper

General · anakin87

Q8_0Excellent
4.8 GB13% of RAM~56 tok/sEstimated3.82B params
Run with ToolPiper

General · redix

Q8_0Excellent
4.8 GB13% of RAM~56 tok/sEstimated3.82B params
Run with ToolPiper

32-core vs 40-core GPU

Apple ties the memory bus to the bin on this chip: 410 GB/s at 32 cores and 546 GB/s at 40. That moves tokens per second. It does not move which models fit, because that is memory, not cores.

ConfigurationMemory bandwidthMemory optionsModels that fit
14-core CPU, 32-core GPU410 GB/s36 GBIdentical
16-core CPU, 40-core GPU546 GB/s48, 64, 128 GBIdentical

On openbuddy zero 56b v21.2 32k the 32-core generates about 11 tok/s and the 40-core about 14 tok/s, a 33% difference. Both hold the model at the same quantization.

Measured on the M4 Max

Nobody has submitted a benchmark on the M4 Max yet, so every speed on this page is the formula estimate rather than a measured run. The estimate is bandwidth-driven and calibrated against chips that do have data, which makes it a good guide and not a promise.

ToolPiper contributes a result anonymously when you run the benchmark, and the leaderboard shows every chip that already has one.

Why unified memory is the number that matters

On a PC the model has to fit in GPU VRAM, which is a separate pool from system RAM and usually the smaller of the two. Apple Silicon has one pool. The M4 Max's 546 GB/s bus is shared by CPU, GPU, and Neural Engine, so a 128 GB machine can hand almost all of that to a model with no copy across a bus.

Apple stopped selling this one, which is exactly why it is interesting. The 16-inch chassis has the most thermal headroom Apple ships in a laptop, so sustained token throughput stays close to the burst figure. A used M4 Max at 128 GB still gives you 546 GB/s and a hard 172B ceiling, and neither number degrades with age the way a battery does.

Common questions

Can the M4 Max MacBook Pro 16" run a 70B model?

Yes, at 128 GB. A 70B model at Q4_K_M needs about 46 GB including an 8K context, and 128 GB of unified memory leaves about 112 GB for weights once macOS takes its share. At 36 GB it does not fit at any quantization worth running.

How much unified memory should I get with the M4 Max MacBook Pro 16"?

Memory is the only spec that changes what you can run at all. 36 GB holds about a 47B model at Q4; 128 GB holds about 172B. It is soldered, so this is a one-time decision, and it is the upgrade worth paying for before core count.

How fast are local LLMs on the M4 Max?

Token generation is bandwidth-bound, so M4 Max throughput scales with its 546 GB/s memory bus. Divide bandwidth by the size of the weights actually read per token to get the ceiling, then expect roughly half of that in practice. A 7B model at Q4 reads about 4 GB per token pass, so the M4 Max lands in the tens of tokens per second and a 70B model lands in the single digits.

Is the 40-core GPU worth it over the 32-core on the M4 Max?

For throughput, yes: Apple ties bandwidth to the bin here, so the 32-core runs at 410 GB/s and the 40-core at 546 GB/s, about 33% more. For fit, no: both bins hold exactly the same models, because that is set by memory rather than by cores.

Is a used M4 Max MacBook Pro 16" still worth buying for local AI?

For inference, the specs that matter do not age: 546 GB/s and up to 128 GB of unified memory are the same numbers today as they were in 2024. A used M4 Max at the top memory option usually beats a new base-tier machine at the same price on both. Check the battery and the display, not the silicon.

Run these models on your MacBook Pro 16"

ToolPiper downloads, manages, and runs local models on Apple Silicon. Free, and nothing leaves the machine.