Which local LLMs can the Apple M4 Max (32-core GPU) run?

Unified memory — by default the GPU can use about 70% of it

Assumes 8k context and an f16 KV cache. Verdicts are shown for each memory size.

Specs

Memory options: 36 GB · 48 GB
Memory bandwidth: 410.0 GB/s
FP16 compute: 27.2 TFLOPS
Launch year: 2024

Verdicts at a glance

At Q4_K_M (or the closest available quant) with 8k context.

36 GBRuns great15 modelsRuns well7 modelsRuns slowly2 modelsWon't run5 models1 more model runs at a lower quant.
48 GBRuns great15 modelsRuns well7 modelsRuns slowly2 modelsWon't run5 models3 more models run at a lower quant.

How to read the verdicts

Runs great: Fully on the GPU at 20 tok/s or more
Runs well: Fully on the GPU at 8–20 tok/s; MoE experts in system RAM at 20 tok/s or more; or at least 90% on the GPU at 8 tok/s or more
Runs slowly: 2–8 tok/s; CPU-only; less than 90% on the GPU; or MoE experts in system RAM below 20 tok/s
Won't run: Does not fit, or under 2 tok/s

Every model on the Apple M4 Max (32-core GPU)

Scroll sideways to see every column.

29 models sorted by verdict and speed, with 48 GB of memory
Model	Verdict	Speed	Quant	Memory	Runs as	Context	Notes
EXAONE 4.0 1.2B	Runs great	est. 187.3 tok/s149.8–224.8calibrated estimate ±20%	Q4_K_M	1.9 / 33.6 GB	Unified memory	up to 64k	—
HyperCLOVA X SEED 1.5B	Runs great	est. 135.5 tok/s108.4–162.6calibrated estimate ±20%	Q4_K_M	2.4 / 33.6 GB	Unified memory	up to 16k	—
Qwen3.5 35B-A3B	Runs great	est. 59.8 tok/s47.9–71.8calibrated estimate ±20%	Q4_K_M	23.4 / 33.6 GB	Unified memory	up to 128k	—
gpt-oss-20b	Runs great	est. 53.8 tok/s43.0–64.5calibrated estimate ±20%	MXFP4	12.9 / 33.6 GB	Unified memory	up to 64k	reasoning model
Qwen3 30B-A3B (2507)	Runs great	est. 43.7 tok/s35.0–52.5calibrated estimate ±20%	Q4_K_M	19.9 / 33.6 GB	Unified memory	up to 32k	—
Gemma 4 26B-A4B	Runs great	est. 42.6 tok/s34.0–51.1calibrated estimate ±20%	Q4_K_M	17.9 / 33.6 GB	Unified memory	up to 64k	—
DeepSeek R1 Distill Llama 8B	Runs great	est. 41.0 tok/s32.8–49.3calibrated estimate ±20%	Q4_K_M	6.6 / 33.6 GB	Unified memory	up to 16k	reasoning model
Kanana 1.5 8B	Runs great	est. 41.0 tok/s32.8–49.3calibrated estimate ±20%	Q4_K_M	6.6 / 33.6 GB	Unified memory	up to 32k	—
Llama 3.1 8B	Runs great	est. 41.0 tok/s32.8–49.3calibrated estimate ±20%	Q4_K_M	6.6 / 33.6 GB	Unified memory	up to 32k	—
Kanana 1.5 15.7B-A3B	Runs great	est. 40.1 tok/s32.1–48.1calibrated estimate ±20%	Q4_K_M	12.1 / 33.6 GB	Unified memory	up to 16k	—
Qwen3.5 9B	Runs great	est. 40.1 tok/s32.1–48.1calibrated estimate ±20%	Q4_K_M	6.7 / 33.6 GB	Unified memory	up to 128k	—
Qwen3 8B	Runs great	est. 39.4 tok/s31.5–47.3calibrated estimate ±20%	Q4_K_M	6.8 / 33.6 GB	Unified memory	up to 32k	—
Gemma 4 12B	Runs great	est. 32.1 tok/s25.7–38.6calibrated estimate ±20%	Q4_K_M	8.2 / 33.6 GB	Unified memory	up to 64k	—
Gemma 3 12B	Runs great	est. 31.4 tok/s25.1–37.7calibrated estimate ±20%	Q4_K_M	8.4 / 33.6 GB	Unified memory	up to 64k	—
Qwen3 14B	Runs great	est. 23.8 tok/s19.0–28.5calibrated estimate ±20%	Q4_K_M	10.9 / 33.6 GB	Unified memory	up to 16k	—
HyperCLOVA X SEED Think 14B	Runs well	est. 24.1 tok/s16.9–31.4theoretical estimate ±30%	Q4_K_M	10.8 / 33.6 GB	Unified memory	up to 64k	reasoning model
Phi-4	Runs well	est. 22.9 tok/s18.3–27.5calibrated estimate ±20%	Q4_K_M	11.3 / 33.6 GB	Unified memory	up to 16k	reasoning model
Mistral Small 3.2 24B	Runs well	est. 15.7 tok/s12.6–18.8calibrated estimate ±20%	Q4_K_M	16.2 / 33.6 GB	Unified memory	up to 64k	Q2_K (heavy quality loss): Runs great
Gemma 3 27B	Runs well	est. 14.3 tok/s11.4–17.1calibrated estimate ±20%	Q4_K_M	17.8 / 33.6 GB	Unified memory	up to 128k	Q2_K (heavy quality loss): Runs great
Qwen3.5 27B	Runs well	est. 13.9 tok/s11.2–16.7calibrated estimate ±20%	Q4_K_M	18.2 / 33.6 GB	Unified memory	up to 128k	—
EXAONE 4.0 32B	Runs well	est. 12.4 tok/s9.9–14.9calibrated estimate ±20%	Q4_K_M	20.5 / 33.6 GB	Unified memory	up to 16k	reasoning model
Qwen3 32B	Runs well	est. 11.2 tok/s9.0–13.5calibrated estimate ±20%	Q4_K_M	22.5 / 33.6 GB	Unified memory	up to 32k	—
EXAONE 4.5 33B	Runs slowly	est. 11.9 tok/s9.6–14.3calibrated estimate ±20%	Q4_K_M	21.2 / 33.6 GB	Unified memory	up to 128k	Try IQ4_XS: Runs well reasoning model
DeepSeek R1 Distill Qwen 32B	Runs slowly	est. 11.2 tok/s8.9–13.4calibrated estimate ±20%	Q4_K_M	22.6 / 33.6 GB	Unified memory	up to 32k	Q2_K (heavy quality loss): Runs well reasoning model
Qwen3.5 122B-A10B	Won't runTry UD-Q2_K_XL (heavy quality loss): Runs slowly	est. 33.9 tok/s23.7–44.1theoretical estimate ±30%	UD-Q2_K_XL	43.1 GB RAM	CPU only	up to 32k	CPU inference — capped at Runs slowly
Solar Open 100B	Won't runTry Q2_K (heavy quality loss): Runs slowly	est. 20.4 tok/s14.3–26.6theoretical estimate ±30%	Q2_K	39.3 GB RAM	CPU only	up to 16k	CPU inference — capped at Runs slowly
Llama 3.3 70B	Won't runTry IQ4_XS: Runs slowly	est. 3.5 tok/s2.5–4.6theoretical estimate ±30%	IQ4_XS	40.6 GB RAM	CPU only	up to 16k	Q2_K (heavy quality loss): Runs well
gpt-oss-120b	Won't run	—	MXFP4	needs 64.3 GB	—	—	Needs about 64.3 GB; 44.0 GB of memory is free (33.6 GB usable by the GPU) reasoning model
Solar Open 2 250B	Won't run	—	IQ4_XS	needs 137.2 GB	—	—	Needs about 137.2 GB; 44.0 GB of memory is free (33.6 GB usable by the GPU)

Reasoning models spend extra tokens thinking, so their speed thresholds are 1.5× stricter (30 / 12 / 3 tok/s).

Verdict by memory size

Q4_K_M verdict for each memory configuration — open the model page for speeds
Model	36 GB	48 GB
EXAONE 4.0 1.2B	Runs great	Runs great
HyperCLOVA X SEED 1.5B	Runs great	Runs great
Qwen3.5 35B-A3B	Runs great	Runs great
gpt-oss-20b	Runs great	Runs great
Qwen3 30B-A3B (2507)	Runs great	Runs great
Gemma 4 26B-A4B	Runs great	Runs great
DeepSeek R1 Distill Llama 8B	Runs great	Runs great
Kanana 1.5 8B	Runs great	Runs great
Llama 3.1 8B	Runs great	Runs great
Kanana 1.5 15.7B-A3B	Runs great	Runs great
Qwen3.5 9B	Runs great	Runs great
Qwen3 8B	Runs great	Runs great
Gemma 4 12B	Runs great	Runs great
Gemma 3 12B	Runs great	Runs great
Qwen3 14B	Runs great	Runs great
HyperCLOVA X SEED Think 14B	Runs well	Runs well
Phi-4	Runs well	Runs well
Mistral Small 3.2 24B	Runs well	Runs well
Gemma 3 27B	Runs well	Runs well
Qwen3.5 27B	Runs well	Runs well
EXAONE 4.0 32B	Runs well	Runs well
Qwen3 32B	Runs well	Runs well
EXAONE 4.5 33B	Runs slowly	Runs slowly
DeepSeek R1 Distill Qwen 32B	Runs slowly	Runs slowly
Qwen3.5 122B-A10B	Won't run	Won't run
Solar Open 100B	Won't run	Won't run
Llama 3.3 70B	Won't run	Won't run
gpt-oss-120b	Won't run	Won't run
Solar Open 2 250B	Won't run	Won't run

Measured results on the Apple M4 Max (32-core GPU)

No public measurements for this device yet.

Frequently asked questions

What is the largest model that runs entirely on the Apple M4 Max (32-core GPU)?

Qwen3.5 35B-A3B at Q4_K_M (a 22.63 GB file) fits entirely in 48 GB of unified memory with 8k context, at est. 59.8 tok/s (47.9–71.8, calibrated estimate ±20%).

How many local LLMs run well on the Apple M4 Max (32-core GPU)?

At Q4_K_M with 8k context, with 48 GB of memory, out of 29 tracked models: 15 run great, 7 run well, 2 run slowly and 5 won't run.

Can the Apple M4 Max (32-core GPU) run a 70B model like Llama 3.3 70B?

Runs slowly — CPU only, IQ4_XS: est. 3.5 tok/s (2.5–4.6, theoretical estimate ±30%).

Speeds are estimates from memory bandwidth, calibrated against public benchmarks, and each one comes with an error band and a confidence label. Real results vary with drivers, backend, context length and thermals.