What are the best models to run on Apple M2 Pro (16GB)?

Based on model rankings and 12 GB usable VRAM: Top Overall: Phi-4 Reasoning Plus (INT4) ~13 tok/s. Top Coding: K2 Horizon 7B (INT4) ~37 tok/s. Optimal Daily Driver: Qwen3.5-9B (INT4) ~29 tok/s. Maximum Model Size: ERNIE-4.5-21B-A3B (INT4) ~55 tok/s.

How are VRAM, throughput, and latency calculated?

Calculations are determined using apxml.com modeling, combining LLM architecture analysis with GPU hardware physics. VRAM and Context Scaling: Memory requirements are determined from model architecture, including parameter count, layer configuration, and attention structure. The KV cache footprint scales deterministically based on context length, batch size, and cache precision. Throughput (TPS): Token generation is calculated in the memory-bandwidth-bound decode phase, measuring the time required to stream model weights and KV cache relative to hardware memory bandwidth. Latency (TTFT): Time to first token is calculated from the compute operations required during prompt prefill relative to hardware tensor compute capacity.

How accurate are these estimates?

Calculations are calibrated against empirical ground-truth benchmark runs across diverse architectures and hardware. Because the underlying mechanics are benchmark-calibrated, estimates remain accurate across different configurations without needing to run physical tests for every scenario. VRAM Accuracy: Memory estimates are deterministic and closely match actual allocation. Minor differences typically stay within 3% to 5% due to CUDA runtime overhead or memory allocator fragmentation. Speed and Latency Accuracy: Throughput and TTFT typically vary within 5% to 10% depending on the specific inference runtime, kernel implementations, and system power limits.

Best LLMs to Run on Apple M2 Pro (16GB)

Rank	Model	Parameters	Precision	TPS (Approx.)	TTFT
#112	Phi-4 Reasoning Plus	14B	INT4	13 tok/s	~29.5s
#125	Qwen3.5-9B	9B	INT4 Q8	29 tok/s 16 tok/s	~16.2s ~17.6s
#128	K2 Horizon 7B	7B	INT4 Q8	37 tok/s 22 tok/s	~12.7s ~12.7s
#135	Gemma 4 12B	11.95B	INT4	15 tok/s	~23.3s
#140	Qwen3.5-4B	4B	INT4 Q8 FP16	57 tok/s 36 tok/s 19 tok/s	~7.4s ~7.4s ~7.5s
#144	GLM-4V	9B	INT4 Q8	21 tok/s 13 tok/s	~16.2s ~19.1s
#150	NVIDIA Nemotron 3 Nano 30B-A3B	3.5B (30B active)	INT4 Q8 FP16	9 tok/s 5 tok/s 3 tok/s	~51.1s ~51.2s ~55.8s
#158	Granite 4.2 8B	8B	INT4 Q8	33 tok/s 19 tok/s	~14.4s ~14.4s
#159	Gemma 4 E4B	8B	INT4 Q8	33 tok/s 19 tok/s	~14.4s ~14.4s
#166	Phi-4	14B	INT4	20 tok/s	~25.0s
#178	OLMo 3 7B Think	7B	INT4 Q8	25 tok/s 17 tok/s	~12.7s ~13.8s
#179	DeepSeek-R1 14B	14B	INT4	20 tok/s	~25.0s
#183	Ministral 3 14B	14B	INT4	13 tok/s	~29.5s
#186	Gemma 3 12B	12B	INT4	23 tok/s	~21.5s
#189	Ministral 3 8B	8B	INT4 Q8	23 tok/s 15 tok/s	~14.4s ~15.7s
#191	MiniCPM5-2B	2B	INT4 Q8 FP16	93 tok/s 62 tok/s 36 tok/s	~3.9s ~3.9s ~4.0s
#192	Qwen3.5-2B	2B	INT4 Q8 FP16	93 tok/s 62 tok/s 36 tok/s	~3.9s ~3.9s ~4.0s
#194	Qwen2.5-14B	14B	INT4	20 tok/s	~25.0s
#200	Gemma 4 E2B	5.1B	INT4 Q8 FP16	47 tok/s 29 tok/s 15 tok/s	~9.3s ~9.4s ~10.2s
#202	DeepSeek-R1 8B	8B	INT4 Q8	33 tok/s 20 tok/s	~14.4s ~14.4s
#204	Qwen3-4B	4B	INT4 Q8 FP16	57 tok/s 36 tok/s 19 tok/s	~7.4s ~7.4s ~7.5s
#205	Ministral 3 3B	3B	INT4 Q8 FP16	45 tok/s 34 tok/s 21 tok/s	~5.7s ~5.7s ~5.7s
#207	Gemma 2 9B	9B	INT4 Q8	29 tok/s 16 tok/s	~16.2s ~17.6s
#208	Qwen3.5-0.8B	0.8B	INT4 Q8 FP16	158 tok/s 119 tok/s 76 tok/s	~1.6s ~1.6s ~1.6s
#209	Qwen3-14B	14B	INT4	20 tok/s	~25.0s
#210	Gemma 3 4B	4B	INT4 Q8 FP16	57 tok/s 36 tok/s 19 tok/s	~7.4s ~7.4s ~7.5s
#211	Qwen3-8B	8B	INT4 Q8	33 tok/s 19 tok/s	~14.4s ~14.4s
#213	DeepSeek-R1 7B	7B	INT4 Q8	38 tok/s 22 tok/s	~12.7s ~12.7s
#214	Qwen3-1.7B	1.7B	INT4 Q8 FP16	103 tok/s 71 tok/s 41 tok/s	~3.4s ~3.4s ~3.4s
#215	Llama 3.1 8B	8B	INT4 Q8	33 tok/s 19 tok/s	~14.4s ~14.4s
#216	Gemma 3 270M	0.27B	INT4 Q8 FP16	199 tok/s 176 tok/s 137 tok/s	~520 ms ~521 ms ~523 ms
#218	Qwen3-0.6B	0.6B	INT4 Q8 FP16	181 tok/s 141 tok/s 94 tok/s	~1.2s ~1.2s ~1.2s
#219	Llama 3 8B	8B	INT4 Q8	33 tok/s 19 tok/s	~14.4s ~14.4s
#221	Phi-4-Mini	3.8B	INT4 Q8 FP16	60 tok/s 37 tok/s 20 tok/s	~7.1s ~7.1s ~7.1s
#223	Qwen2.5-7B	7B	INT4 Q8	37 tok/s 22 tok/s	~12.7s ~12.7s
#224	Ministral-8B-2410	8B	INT4 Q8	33 tok/s 19 tok/s	~14.4s ~14.4s
#225	DeepSeek-R1 1.5B	1.5B	INT4 Q8 FP16	112 tok/s 78 tok/s 46 tok/s	~3.0s ~3.0s ~3.0s
#226	Llama 3.2 3B	3B	INT4 Q8 FP16	71 tok/s 45 tok/s 25 tok/s	~5.7s ~5.7s ~5.7s
#228	Phi-3-mini	3.8B	INT4 Q8 FP16	60 tok/s 37 tok/s 20 tok/s	~7.1s ~7.1s ~7.1s
#229	Gemma 3n E2B IT	6B (2B active)	INT4 Q8	45 tok/s 37 tok/s	~4.2s ~4.2s
#230	Mistral-7B-Instruct-v0.1	7.3B	INT4 Q8	35 tok/s 21 tok/s	~13.2s ~13.2s
#231	Llama 3.2 1B	1B	INT4 Q8 FP16	140 tok/s 102 tok/s 64 tok/s	~2.0s ~2.0s ~2.0s
#232	Gemma 2 2B	2B	INT4 Q8 FP16	93 tok/s 62 tok/s 36 tok/s	~3.9s ~3.9s ~4.0s
#233	Phi-3-medium	14B	INT4	20 tok/s	~25.0s
#234	OLMo 3 7B Instruct	7B	INT4 Q8	25 tok/s 17 tok/s	~12.7s ~13.8s
#236	Qwen2-7B	7B	INT4 Q8	37 tok/s 22 tok/s	~12.7s ~12.7s
#237	Phi-3-small	7B	INT4 Q8	37 tok/s 22 tok/s	~12.7s ~12.7s
#238	Gemma 3 1B	1B	INT4 Q8 FP16	140 tok/s 102 tok/s 64 tok/s	~2.0s ~2.0s ~2.0s
#239	Mistral-7B-Instruct-v0.2	7.3B	INT4 Q8	35 tok/s 21 tok/s	~13.2s ~13.2s
#241	Gemma 1 7B	7B	INT4 Q8	25 tok/s 17 tok/s	~12.7s ~13.8s
#242	Gemma 1 2B	2B	INT4 Q8 FP16	100 tok/s 65 tok/s 37 tok/s	~3.9s ~3.9s ~4.0s
#243	ChatGLM3-6B	6B	INT4 Q8	28 tok/s 20 tok/s	~10.9s ~10.9s
#244	ChatGLM2-6B	6B	INT4 Q8	28 tok/s 20 tok/s	~10.9s ~10.9s
#245	ChatGLM-6B	6B	INT4 Q8	28 tok/s 20 tok/s	~10.9s ~10.9s
Unranked Models (Sorted by Speed)
-	DINOv3 ViT-L/16	0.3B	INT4 Q8 FP16	228 tok/s 193 tok/s 144 tok/s	~579 ms ~580 ms ~582 ms
-	Qwen2-0.5B	0.5B	INT4 Q8 FP16	195 tok/s 155 tok/s 106 tok/s	~972 ms ~973 ms ~976 ms
-	Qwen2.5-0.5B	0.5B	INT4 Q8 FP16	195 tok/s 155 tok/s 106 tok/s	~972 ms ~973 ms ~976 ms
-	ERNIE-4.5-0.3B	0.3B	INT4 Q8 FP16	189 tok/s 167 tok/s 129 tok/s	~580 ms ~581 ms ~583 ms
-	ERNIE-4.5-0.3B-Base	0.3B	INT4 Q8 FP16	189 tok/s 167 tok/s 129 tok/s	~580 ms ~581 ms ~583 ms
-	Liquid D1-Omni 600M	0.6B	INT4 Q8 FP16	181 tok/s 141 tok/s 94 tok/s	~1.2s ~1.2s ~1.2s
-	PixArt-α XL 2 1024-MS	0.61B	INT4 Q8 FP16	180 tok/s 140 tok/s 93 tok/s	~1.2s ~1.2s ~1.2s
-	PixArt-Σ XL 2 1024-MS	0.61B	INT4 Q8 FP16	180 tok/s 140 tok/s 93 tok/s	~1.2s ~1.2s ~1.2s
-	Nemotron 3.5 Parse	0.63B	INT4 Q8 FP16	177 tok/s 137 tok/s 91 tok/s	~1.2s ~1.2s ~1.2s
-	SD 2.1 Turbo	0.86B	INT4 Q8 FP16	153 tok/s 113 tok/s 72 tok/s	~1.7s ~1.7s ~1.7s
-	Stable Diffusion 1.4	0.86B	INT4 Q8 FP16	153 tok/s 113 tok/s 72 tok/s	~1.7s ~1.7s ~1.7s
-	Stable Diffusion 2.1	0.86B	INT4 Q8 FP16	153 tok/s 113 tok/s 72 tok/s	~1.7s ~1.7s ~1.7s
-	Falcon-1B	1B	INT4 Q8 FP16	150 tok/s 107 tok/s 66 tok/s	~2.0s ~2.0s ~2.0s
-	Falcon3-1B	1B	INT4 Q8 FP16	140 tok/s 102 tok/s 64 tok/s	~2.0s ~2.0s ~2.0s
-	MiniCPM-V 4.7 1B	1B	INT4 Q8 FP16	140 tok/s 102 tok/s 64 tok/s	~2.0s ~2.0s ~2.0s
-	Kandinsky 2.2 Decoder	1.22B	INT4 Q8 FP16	126 tok/s 90 tok/s 55 tok/s	~2.4s ~2.4s ~2.5s
-	Wan 2.1 T2V 1.3B	1.3B	INT4 Q8 FP16	122 tok/s 86 tok/s 52 tok/s	~2.6s ~2.6s ~2.6s
-	Qwen2-1.5B	1.5B	INT4 Q8 FP16	112 tok/s 78 tok/s 46 tok/s	~3.0s ~3.0s ~3.0s
-	Qwen2.5-1.5B	1.5B	INT4 Q8 FP16	112 tok/s 78 tok/s 46 tok/s	~3.0s ~3.0s ~3.0s
-	SVD Image-to-Video XT	1.5B	INT4 Q8 FP16	112 tok/s 78 tok/s 46 tok/s	~3.0s ~3.0s ~3.0s
-	SVD Image-to-Video XT 1.1	1.5B	INT4 Q8 FP16	112 tok/s 78 tok/s 46 tok/s	~3.0s ~3.0s ~3.0s
-	LFM3 1.7B	1.7B	INT4 Q8 FP16	103 tok/s 71 tok/s 41 tok/s	~3.4s ~3.4s ~3.4s
-	LFM3 1.7B Base	1.7B	INT4 Q8 FP16	103 tok/s 71 tok/s 41 tok/s	~3.4s ~3.4s ~3.4s
-	Spark X2.5 1.7B	1.7B	INT4 Q8 FP16	103 tok/s 71 tok/s 41 tok/s	~3.4s ~3.4s ~3.4s
-	Hy-MT2 1.8B	1.8B	INT4 Q8 FP16	100 tok/s 68 tok/s 39 tok/s	~3.5s ~3.5s ~3.6s
-	CogVideoX 2B	2B	INT4 Q8 FP16	93 tok/s 62 tok/s 36 tok/s	~3.9s ~3.9s ~4.0s
-	LFM3 Audio 2B	2B	INT4 Q8 FP16	93 tok/s 62 tok/s 36 tok/s	~3.9s ~3.9s ~4.0s
-	LFM3 Omni 2B	2B	INT4 Q8 FP16	93 tok/s 62 tok/s 36 tok/s	~3.9s ~3.9s ~4.0s
-	LFM3 VL 2B	2B	INT4 Q8 FP16	93 tok/s 62 tok/s 36 tok/s	~3.9s ~3.9s ~4.0s
-	Liquid D1 2B	2B	INT4 Q8 FP16	93 tok/s 62 tok/s 36 tok/s	~3.9s ~3.9s ~4.0s
-	LTX-Video 0.9.1	2B	INT4 Q8 FP16	93 tok/s 62 tok/s 36 tok/s	~3.9s ~3.9s ~4.0s
-	Moondream 3.1	9B (2B active)	INT4 Q8	93 tok/s 59 tok/s	~4.4s ~4.8s
-	Stable Diffusion 3 Medium	2B	INT4 Q8 FP16	93 tok/s 62 tok/s 36 tok/s	~3.9s ~3.9s ~4.0s
-	Kandinsky 3.0	2.1B	INT4 Q8 FP16	90 tok/s 60 tok/s 34 tok/s	~4.1s ~4.1s ~4.1s
-	LTX-2	2.2B	INT4 Q8 FP16	88 tok/s 58 tok/s 33 tok/s	~4.3s ~4.3s ~4.3s
-	SDXL 1.0 Refiner	2.31B	INT4 Q8 FP16	85 tok/s 56 tok/s 32 tok/s	~4.5s ~4.5s ~4.5s
-	Stable Diffusion 3.5 Medium	2.5B	INT4 Q8 FP16	81 tok/s 53 tok/s 30 tok/s	~4.8s ~4.8s ~4.8s
-	SDXL 1.0 Base	2.56B	INT4 Q8 FP16	79 tok/s 52 tok/s 29 tok/s	~4.9s ~4.9s ~4.9s
-	SDXL Turbo	2.56B	INT4 Q8 FP16	79 tok/s 52 tok/s 29 tok/s	~4.9s ~4.9s ~4.9s
-	Falcon-3B	3B	INT4 Q8 FP16	76 tok/s 47 tok/s 26 tok/s	~5.7s ~5.7s ~5.7s
-	CroissantLLM Base	1.3B	INT4 Q8 FP16	74 tok/s 61 tok/s 41 tok/s	~2.6s ~2.6s ~2.6s
-	Phi-1	1.3B	INT4 Q8 FP16	74 tok/s 61 tok/s 41 tok/s	~2.6s ~2.6s ~2.6s
-	Phi-1.5	1.3B	INT4 Q8 FP16	74 tok/s 61 tok/s 41 tok/s	~2.6s ~2.6s ~2.6s
-	DeepSeek-R1 3B	3B	INT4 Q8 FP16	72 tok/s 46 tok/s 25 tok/s	~5.7s ~5.7s ~5.7s
-	Mellum 2.1 12B A2.5B Thinking	12B (2.5B active)	INT4	72 tok/s	~5.5s
-	Falcon3-3B	3B	INT4 Q8 FP16	71 tok/s 45 tok/s 25 tok/s	~5.7s ~5.7s ~5.7s
-	LFM2-VL 3B	3B	INT4 Q8 FP16	71 tok/s 45 tok/s 25 tok/s	~5.7s ~5.7s ~5.7s
-	Liquid D1 3B	3B	INT4 Q8 FP16	71 tok/s 45 tok/s 25 tok/s	~5.7s ~5.7s ~5.7s
-	Liquid D1-VL 3B	3B	INT4 Q8 FP16	71 tok/s 45 tok/s 25 tok/s	~5.7s ~5.7s ~5.7s
-	Liquid LFM2-VL 3B D1	3B	INT4 Q8 FP16	71 tok/s 45 tok/s 25 tok/s	~5.7s ~5.7s ~5.7s
-	Qwen2.5-3B	3B	INT4 Q8 FP16	71 tok/s 45 tok/s 25 tok/s	~5.7s ~5.7s ~5.7s
-	FLUX.2 Klein 4B	4B	INT4 Q8 FP16	57 tok/s 36 tok/s 19 tok/s	~7.4s ~7.4s ~7.5s
-	FrogNano 4B	4B	INT4 Q8 FP16	57 tok/s 36 tok/s 19 tok/s	~7.4s ~7.4s ~7.5s
-	Mage-VL Cowork	4B	INT4 Q8 FP16	57 tok/s 36 tok/s 19 tok/s	~7.4s ~7.4s ~7.5s
-	Spark X2.5 4B	4B	INT4 Q8 FP16	57 tok/s 36 tok/s 19 tok/s	~7.4s ~7.4s ~7.5s
-	EmbeddingGemma 2	2B	INT4 Q8 FP16	55 tok/s 44 tok/s 29 tok/s	~3.9s ~3.9s ~4.0s
-	ERNIE-4.5-21B-A3B	21B (3B active)	INT4	55 tok/s	~8.2s
-	ERNIE-4.5-21B-A3B-Base	21B (3B active)	INT4	55 tok/s	~8.2s
-	CogVideoX 1.5 5B	5B	INT4 Q8 FP16	48 tok/s 29 tok/s 15 tok/s	~9.2s ~9.2s ~10.0s
-	CogVideoX 5B	5B	INT4 Q8 FP16	48 tok/s 29 tok/s 15 tok/s	~9.2s ~9.2s ~10.0s
-	Wan 2.2 TI2V 5B	5B	INT4 Q8 FP16	48 tok/s 29 tok/s 15 tok/s	~9.2s ~9.2s ~10.0s
-	Phi-2	2.7B	INT4 Q8 FP16	47 tok/s 36 tok/s 23 tok/s	~5.2s ~5.2s ~5.2s
-	MaLLaM-3B	3B	INT4 Q8 FP16	45 tok/s 34 tok/s 21 tok/s	~5.7s ~5.7s ~5.7s
-	SmolLM3 3B	3B	INT4 Q8 FP16	45 tok/s 34 tok/s 21 tok/s	~5.7s ~5.7s ~5.7s
-	Falcon-7B	7B	INT4 Q8	38 tok/s 22 tok/s	~12.7s ~12.7s
-	AuraFlow v0.2	6.8B	INT4 Q8	37 tok/s 22 tok/s	~12.3s ~12.3s
-	AuraFlow v0.3	6.8B	INT4 Q8	37 tok/s 22 tok/s	~12.3s ~12.3s
-	Falcon3-7B	7B	INT4 Q8	37 tok/s 22 tok/s	~12.7s ~12.7s
-	Hy-MT2 7B	7B	INT4 Q8	37 tok/s 22 tok/s	~12.7s ~12.7s
-	Mistral-7B-v0.1	7.3B	INT4 Q8	35 tok/s 21 tok/s	~13.2s ~13.2s
-	Youtu Parsing Omni	8B	INT4 Q8	33 tok/s 20 tok/s	~14.4s ~14.4s
-	Stable Diffusion 3.5 Large	8.1B	INT4 Q8	32 tok/s 19 tok/s	~14.6s ~14.6s
-	Stable Diffusion 3.5 Large Turbo	8.1B	INT4 Q8	32 tok/s 19 tok/s	~14.6s ~14.6s
-	FLUX.2 Klein 9B	9B	INT4 Q8	29 tok/s 16 tok/s	~16.2s ~17.6s
-	LensVLM 9B	9B	INT4 Q8	29 tok/s 16 tok/s	~16.2s ~17.6s
-	MiMo V2.6 Distill Qwen 9B	9B	INT4 Q8	29 tok/s 16 tok/s	~16.2s ~17.6s
-	ChatGLM3-6B-32K	6B	INT4 Q8	28 tok/s 20 tok/s	~10.9s ~10.9s
-	Yi-6B	6B	INT4 Q8	28 tok/s 20 tok/s	~10.9s ~10.9s
-	Falcon3-10B	10B	INT4 Q8	27 tok/s 14 tok/s	~18.0s ~21.2s
-	Mochi 1 Preview	10B	INT4 Q8	27 tok/s 14 tok/s	~18.0s ~21.2s
-	Falcon2-11B	11B	INT4 Q8	26 tok/s 13 tok/s	~19.7s ~23.2s
-	Hunyuan Lite	7B	INT4 Q8	25 tok/s 17 tok/s	~12.7s ~13.8s
-	OLMo 3 7B Base	7B	INT4 Q8	25 tok/s 17 tok/s	~12.7s ~13.8s
-	SEA-LION-7B	7.1B	INT4 Q8	25 tok/s 16 tok/s	~12.9s ~14.0s
-	SEA-LION-7B-Instruct	7.1B	INT4 Q8	25 tok/s 16 tok/s	~12.9s ~14.0s
-	FLUX.1 Canny Dev	11.9B	INT4	23 tok/s	~21.3s
-	FLUX.1 Depth Dev	11.9B	INT4	23 tok/s	~21.3s
-	FLUX.1 Dev	11.9B	INT4	23 tok/s	~21.3s
-	FLUX.1 Fill Dev	11.9B	INT4	23 tok/s	~21.3s
-	FLUX.1 Schnell	11.9B	INT4	23 tok/s	~21.3s
-	Sahabat-AI-Llama3-8B-Instruct	8B	INT4 Q8	23 tok/s 15 tok/s	~14.4s ~15.7s
-	Typhoon-2-8B	8B	INT4 Q8	23 tok/s 15 tok/s	~14.4s ~15.7s
-	GLM-4-9B	9B	INT4 Q8	21 tok/s 13 tok/s	~16.2s ~19.1s
-	GLM-4-9B-Chat	9B	INT4 Q8	21 tok/s 13 tok/s	~16.2s ~19.1s
-	GLM-4-9B-Chat-1M	9B	INT4 Q8	21 tok/s 13 tok/s	~16.2s ~19.1s
-	HunyuanVideo 1.5 720P T2V	13B	INT4	21 tok/s	~23.3s
-	Yi-9B	9B	INT4 Q8	21 tok/s 13 tok/s	~16.2s ~19.1s
-	Sahabat-AI-Gemma2-9B	9.2B	INT4	20 tok/s	~16.6s
-	Sahabat-AI-Gemma2-9B-Instruct	9.2B	INT4	20 tok/s	~16.6s
-	Wan 2.1 I2V 14B 720P	14B	INT4	20 tok/s	~25.0s
-	Wan 2.1 T2V 14B	14B	INT4	20 tok/s	~25.0s
-	Wan 2.2 T2V A14B	14B	INT4	20 tok/s	~25.0s
-	FLUX.2 Dev	15B	INT4	18 tok/s	~26.7s
-	GreenMind-14B-R1	14B	INT4	13 tok/s	~29.5s

Apple M2 Pro (16GB)

VRAM Scaling by Context Length

Best Models to Run on Apple M2 Pro (16GB)

Inference Capacity

Fine-Tuning Capacity

Workload Recommendations

Frequently Asked Questions