Apple M3 Max (36GB)
The Apple M3 Max (36GB) uses a unified memory architecture. For reliable local LLM inference without OS memory pressure, this benchmark models a 75% usable allocation (27 GB) with 300 GB/s unified memory bandwidth.
Memory
36 GB
Usable Memory Ceiling
27 GB (75% usable budget)
Bandwidth
300 GB/s
Max Inference Class
32B Models (Q4)
TDP
115W
VRAM Scaling by Context Length
Estimated memory requirements as context length increases. Curves crossing above the reference line indicate Out of Memory (OOM).
Precision:
Best Models to Run on Apple M3 Max (36GB)
Inference throughput, memory compatibility, and generation speeds for open-weights LLMs on Apple M3 Max (36GB).
Sort by:
Context:
KV Cache:
| Rank | Model | Parameters | Precision | TPS (Approx.) | TTFT | Can Run |
|---|---|---|---|---|---|---|
#48 | Qwen 3.8 27B | 27B | INT4 | 16 tok/s | ~25.4s | |
#83 | K2 Horizon MoVA 36B A4B | 36B (4B active) | INT4 | 66 tok/s | ~5.0s | |
#93 | Qwen3.5-27B | 27B | INT4 | 16 tok/s | ~25.4s | |
#105 | Sarvam-30B | 32B (2.4B active) | INT4 | 87 tok/s | ~3.5s | |
#106 | Qwen3.6 35B A3B | 35B (3B active) | INT4 | 78 tok/s | ~4.0s | |
#110 | Muse Glimmer 30B | 30B | INT4 | 14 tok/s | ~28.2s | |
#111 | MiMo V2 Flash | 15B (309B active) | INT4 Q8 | 1 tok/s 1 tok/s | ~278.0s ~278.6s | |
#112 | Phi-4 Reasoning Plus | 14B | INT4 Q8 | 21 tok/s 14 tok/s | ~13.4s ~13.4s | |
#117 | Gemma 4 26B A4B | 25.2B (3.8B active) | INT4 Q8 | 71 tok/s 44 tok/s | ~4.5s ~5.4s | |
#125 | Qwen3.5-9B | 9B | INT4 Q8 FP16 | 42 tok/s 25 tok/s 13 tok/s | ~8.7s ~8.7s ~8.7s | |
#126 | Qwen3.5-35B-A3B | 35B (3B active) | INT4 | 78 tok/s | ~4.0s | |
#128 | K2 Horizon 7B | 7B | INT4 Q8 FP16 | 52 tok/s 31 tok/s 17 tok/s | ~6.8s ~6.8s ~6.8s | |
#135 | Gemma 4 12B | 11.95B | INT4 Q8 FP16 | 24 tok/s 16 tok/s 8 tok/s | ~11.4s ~11.5s ~13.5s | |
#140 | Qwen3.5-4B | 4B | INT4 Q8 FP16 | 79 tok/s 51 tok/s 28 tok/s | ~4.0s ~4.0s ~4.0s | |
#143 | GPT-OSS 20B | 21B (3.6B active) | INT4 Q8 | 33 tok/s 26 tok/s | ~4.3s ~5.0s | |
#144 | GLM-4V | 9B | INT4 Q8 FP16 | 30 tok/s 21 tok/s 12 tok/s | ~8.7s ~8.7s ~8.7s | |
#150 | NVIDIA Nemotron 3 Nano 30B-A3B | 3.5B (30B active) | INT4 Q8 FP16 | 14 tok/s 8 tok/s 4 tok/s | ~27.3s ~27.3s ~27.4s | |
#153 | Qwen3-30B-A3B | 30B (3B active) | INT4 | 79 tok/s | ~4.0s | |
#158 | Granite 4.2 8B | 8B | INT4 Q8 FP16 | 46 tok/s 28 tok/s 15 tok/s | ~7.7s ~7.7s ~7.7s | |
#159 | Gemma 4 E4B | 8B | INT4 Q8 FP16 | 46 tok/s 28 tok/s 15 tok/s | ~7.7s ~7.7s ~7.7s | |
#165 | GLM-4 | 32B | INT4 | 10 tok/s | ~32.6s | |
#166 | Phi-4 | 14B | INT4 Q8 | 29 tok/s 17 tok/s | ~13.4s ~13.4s | |
#167 | Gemma 3 27B | 27B | INT4 | 16 tok/s | ~25.4s | |
#170 | DeepSeek-R1 32B | 32B | INT4 | 14 tok/s | ~30.0s | |
#171 | Qwen3-32B | 32B | INT4 | 13 tok/s | ~30.0s | |
#175 | Magistral Small | 24B | INT4 | 13 tok/s | ~22.7s | |
#176 | OLMo 3.1 32B Instruct | 32B | INT4 | 10 tok/s | ~32.6s | |
#177 | Qwen2.5-32B | 32B | INT4 | 13 tok/s | ~30.0s | |
#178 | OLMo 3 7B Think | 7B | INT4 Q8 FP16 | 37 tok/s 26 tok/s 15 tok/s | ~6.8s ~6.8s ~6.8s | |
#179 | DeepSeek-R1 14B | 14B | INT4 Q8 | 29 tok/s 17 tok/s | ~13.4s ~13.4s | |
#183 | Ministral 3 14B | 14B | INT4 Q8 | 21 tok/s 14 tok/s | ~13.4s ~13.4s | |
#186 | Gemma 3 12B | 12B | INT4 Q8 FP16 | 33 tok/s 19 tok/s 10 tok/s | ~11.5s ~11.5s ~12.6s | |
#188 | Mistral-Small-2501 | 24B | INT4 Q8 | 18 tok/s 9 tok/s | ~22.6s ~26.7s | |
#189 | Ministral 3 8B | 8B | INT4 Q8 FP16 | 33 tok/s 23 tok/s 13 tok/s | ~7.7s ~7.7s ~7.8s | |
#190 | Gemma 4 31B | 30.7B | INT4 | 14 tok/s | ~28.8s | |
#191 | MiniCPM5-2B | 2B | INT4 Q8 FP16 | 121 tok/s 85 tok/s 51 tok/s | ~2.1s ~2.1s ~2.1s | |
#192 | Qwen3.5-2B | 2B | INT4 Q8 FP16 | 121 tok/s 85 tok/s 51 tok/s | ~2.1s ~2.1s ~2.1s | |
#193 | Gemma 2 27B | 27B | INT4 | 16 tok/s | ~25.4s | |
#194 | Qwen2.5-14B | 14B | INT4 Q8 | 29 tok/s 17 tok/s | ~13.4s ~13.4s | |
#195 | OLMo 3.1 32B Think | 32B | INT4 | 10 tok/s | ~32.6s | |
#196 | Nemotron 3.5 Lightning | 30B (3B active) | INT4 | 28 tok/s | ~4.3s | |
#200 | Gemma 4 E2B | 5.1B | INT4 Q8 FP16 | 66 tok/s 41 tok/s 23 tok/s | ~5.0s ~5.0s ~5.0s | |
#202 | DeepSeek-R1 8B | 8B | INT4 Q8 FP16 | 48 tok/s 28 tok/s 15 tok/s | ~7.7s ~7.7s ~7.7s | |
#204 | Qwen3-4B | 4B | INT4 Q8 FP16 | 79 tok/s 51 tok/s 28 tok/s | ~4.0s ~4.0s ~4.0s | |
#205 | Ministral 3 3B | 3B | INT4 Q8 FP16 | 63 tok/s 48 tok/s 31 tok/s | ~3.0s ~3.0s ~3.1s | |
#206 | OLMo 3 32B Base | 32B | INT4 | 10 tok/s | ~32.6s | |
#207 | Gemma 2 9B | 9B | INT4 Q8 FP16 | 42 tok/s 25 tok/s 13 tok/s | ~8.7s ~8.7s ~8.7s | |
#208 | Qwen3.5-0.8B | 0.8B | INT4 Q8 FP16 | 188 tok/s 149 tok/s 101 tok/s | ~854 ms ~855 ms ~858 ms | |
#209 | Qwen3-14B | 14B | INT4 Q8 | 29 tok/s 17 tok/s | ~13.4s ~13.4s | |
#210 | Gemma 3 4B | 4B | INT4 Q8 FP16 | 79 tok/s 51 tok/s 28 tok/s | ~4.0s ~4.0s ~4.0s | |
#211 | Qwen3-8B | 8B | INT4 Q8 FP16 | 46 tok/s 28 tok/s 15 tok/s | ~7.7s ~7.7s ~7.7s | |
#213 | DeepSeek-R1 7B | 7B | INT4 Q8 FP16 | 53 tok/s 32 tok/s 17 tok/s | ~6.8s ~6.8s ~6.8s | |
#214 | Qwen3-1.7B | 1.7B | INT4 Q8 FP16 | 132 tok/s 95 tok/s 58 tok/s | ~1.8s ~1.8s ~1.8s | |
#215 | Llama 3.1 8B | 8B | INT4 Q8 FP16 | 46 tok/s 28 tok/s 15 tok/s | ~7.7s ~7.7s ~7.7s | |
#216 | Gemma 3 270M | 0.27B | INT4 Q8 FP16 | 224 tok/s 204 tok/s 168 tok/s | ~282 ms ~282 ms ~283 ms | |
#217 | Command R | 35B | INT4 | 9 tok/s | ~35.6s | |
#218 | Qwen3-0.6B | 0.6B | INT4 Q8 FP16 | 209 tok/s 171 tok/s 122 tok/s | ~631 ms ~632 ms ~634 ms | |
#219 | Llama 3 8B | 8B | INT4 Q8 FP16 | 46 tok/s 28 tok/s 15 tok/s | ~7.7s ~7.7s ~7.7s | |
#221 | Phi-4-Mini | 3.8B | INT4 Q8 FP16 | 82 tok/s 53 tok/s 30 tok/s | ~3.8s ~3.8s ~3.8s | |
#222 | Mixtral-8x7B-v0.1 | 46.7B (7B active) | INT4 | 41 tok/s | ~9.2s | |
#223 | Qwen2.5-7B | 7B | INT4 Q8 FP16 | 52 tok/s 31 tok/s 17 tok/s | ~6.8s ~6.8s ~6.8s | |
#224 | Ministral-8B-2410 | 8B | INT4 Q8 FP16 | 46 tok/s 28 tok/s 15 tok/s | ~7.7s ~7.7s ~7.7s | |
#225 | DeepSeek-R1 1.5B | 1.5B | INT4 Q8 FP16 | 142 tok/s 103 tok/s 64 tok/s | ~1.6s ~1.6s ~1.6s | |
#226 | Llama 3.2 3B | 3B | INT4 Q8 FP16 | 95 tok/s 63 tok/s 36 tok/s | ~3.0s ~3.0s ~3.0s | |
#227 | Nemotron 3.5 Lightning 30B A3B | 30B (3B active) | INT4 | 79 tok/s | ~4.0s | |
#228 | Phi-3-mini | 3.8B | INT4 Q8 FP16 | 82 tok/s 53 tok/s 30 tok/s | ~3.8s ~3.8s ~3.8s | |
#229 | Gemma 3n E2B IT | 6B (2B active) | INT4 Q8 FP16 | 62 tok/s 53 tok/s 37 tok/s | ~2.3s ~2.3s ~2.3s | |
#230 | Mistral-7B-Instruct-v0.1 | 7.3B | INT4 Q8 FP16 | 50 tok/s 30 tok/s 16 tok/s | ~7.0s ~7.1s ~7.1s | |
#231 | Llama 3.2 1B | 1B | INT4 Q8 FP16 | 171 tok/s 131 tok/s 87 tok/s | ~1.1s ~1.1s ~1.1s | |
#232 | Gemma 2 2B | 2B | INT4 Q8 FP16 | 121 tok/s 85 tok/s 51 tok/s | ~2.1s ~2.1s ~2.1s | |
#233 | Phi-3-medium | 14B | INT4 Q8 | 29 tok/s 17 tok/s | ~13.4s ~13.4s | |
#234 | OLMo 3 7B Instruct | 7B | INT4 Q8 FP16 | 37 tok/s 26 tok/s 15 tok/s | ~6.8s ~6.8s ~6.8s | |
#235 | Yi-34B | 34B | INT4 | 9 tok/s | ~34.6s | |
#236 | Qwen2-7B | 7B | INT4 Q8 FP16 | 52 tok/s 31 tok/s 17 tok/s | ~6.8s ~6.8s ~6.8s | |
#237 | Phi-3-small | 7B | INT4 Q8 FP16 | 52 tok/s 31 tok/s 17 tok/s | ~6.8s ~6.8s ~6.8s | |
#238 | Gemma 3 1B | 1B | INT4 Q8 FP16 | 171 tok/s 131 tok/s 87 tok/s | ~1.1s ~1.1s ~1.1s | |
#239 | Mistral-7B-Instruct-v0.2 | 7.3B | INT4 Q8 FP16 | 50 tok/s 30 tok/s 16 tok/s | ~7.0s ~7.1s ~7.1s | |
#241 | Gemma 1 7B | 7B | INT4 Q8 FP16 | 37 tok/s 26 tok/s 15 tok/s | ~6.8s ~6.8s ~6.8s | |
#242 | Gemma 1 2B | 2B | INT4 Q8 FP16 | 129 tok/s 88 tok/s 52 tok/s | ~2.1s ~2.1s ~2.1s | |
#243 | ChatGLM3-6B | 6B | INT4 Q8 FP16 | 41 tok/s 29 tok/s 17 tok/s | ~5.8s ~5.9s ~5.9s | |
#244 | ChatGLM2-6B | 6B | INT4 Q8 FP16 | 41 tok/s 29 tok/s 17 tok/s | ~5.8s ~5.9s ~5.9s | |
#245 | ChatGLM-6B | 6B | INT4 Q8 FP16 | 41 tok/s 29 tok/s 17 tok/s | ~5.8s ~5.9s ~5.9s | |
Unranked Models (Sorted by Speed) | ||||||
- | DINOv3 ViT-L/16 | 0.3B | INT4 Q8 FP16 | 248 tok/s 220 tok/s 174 tok/s | ~313 ms ~314 ms ~315 ms | |
- | Qwen2-0.5B | 0.5B | INT4 Q8 FP16 | 221 tok/s 185 tok/s 136 tok/s | ~523 ms ~524 ms ~526 ms | |
- | Qwen2.5-0.5B | 0.5B | INT4 Q8 FP16 | 221 tok/s 185 tok/s 136 tok/s | ~523 ms ~524 ms ~526 ms | |
- | ERNIE-4.5-0.3B | 0.3B | INT4 Q8 FP16 | 217 tok/s 196 tok/s 159 tok/s | ~314 ms ~314 ms ~315 ms | |
- | ERNIE-4.5-0.3B-Base | 0.3B | INT4 Q8 FP16 | 217 tok/s 196 tok/s 159 tok/s | ~314 ms ~314 ms ~315 ms | |
- | Liquid D1-Omni 600M | 0.6B | INT4 Q8 FP16 | 209 tok/s 171 tok/s 122 tok/s | ~631 ms ~632 ms ~634 ms | |
- | PixArt-α XL 2 1024-MS | 0.61B | INT4 Q8 FP16 | 208 tok/s 170 tok/s 121 tok/s | ~641 ms ~642 ms ~644 ms | |
- | PixArt-Σ XL 2 1024-MS | 0.61B | INT4 Q8 FP16 | 208 tok/s 170 tok/s 121 tok/s | ~641 ms ~642 ms ~644 ms | |
- | Nemotron 3.5 Parse | 0.63B | INT4 Q8 FP16 | 206 tok/s 167 tok/s 118 tok/s | ~666 ms ~667 ms ~669 ms | |
- | SD 2.1 Turbo | 0.86B | INT4 Q8 FP16 | 183 tok/s 143 tok/s 97 tok/s | ~921 ms ~923 ms ~926 ms | |
- | Stable Diffusion 1.4 | 0.86B | INT4 Q8 FP16 | 183 tok/s 143 tok/s 97 tok/s | ~921 ms ~923 ms ~926 ms | |
- | Stable Diffusion 2.1 | 0.86B | INT4 Q8 FP16 | 183 tok/s 143 tok/s 97 tok/s | ~921 ms ~923 ms ~926 ms | |
- | Falcon-1B | 1B | INT4 Q8 FP16 | 181 tok/s 137 tok/s 89 tok/s | ~1.1s ~1.1s ~1.1s | |
- | Falcon3-1B | 1B | INT4 Q8 FP16 | 171 tok/s 131 tok/s 87 tok/s | ~1.1s ~1.1s ~1.1s | |
- | MiniCPM-V 4.7 1B | 1B | INT4 Q8 FP16 | 171 tok/s 131 tok/s 87 tok/s | ~1.1s ~1.1s ~1.1s | |
- | Kandinsky 2.2 Decoder | 1.22B | INT4 Q8 FP16 | 156 tok/s 117 tok/s 75 tok/s | ~1.3s ~1.3s ~1.3s | |
- | Wan 2.1 T2V 1.3B | 1.3B | INT4 Q8 FP16 | 152 tok/s 113 tok/s 72 tok/s | ~1.4s ~1.4s ~1.4s | |
- | Qwen2-1.5B | 1.5B | INT4 Q8 FP16 | 142 tok/s 103 tok/s 64 tok/s | ~1.6s ~1.6s ~1.6s | |
- | Qwen2.5-1.5B | 1.5B | INT4 Q8 FP16 | 142 tok/s 103 tok/s 64 tok/s | ~1.6s ~1.6s ~1.6s | |
- | SVD Image-to-Video XT | 1.5B | INT4 Q8 FP16 | 142 tok/s 103 tok/s 64 tok/s | ~1.6s ~1.6s ~1.6s | |
- | SVD Image-to-Video XT 1.1 | 1.5B | INT4 Q8 FP16 | 142 tok/s 103 tok/s 64 tok/s | ~1.6s ~1.6s ~1.6s | |
- | LFM3 1.7B | 1.7B | INT4 Q8 FP16 | 132 tok/s 95 tok/s 58 tok/s | ~1.8s ~1.8s ~1.8s | |
- | LFM3 1.7B Base | 1.7B | INT4 Q8 FP16 | 132 tok/s 95 tok/s 58 tok/s | ~1.8s ~1.8s ~1.8s | |
- | Spark X2.5 1.7B | 1.7B | INT4 Q8 FP16 | 132 tok/s 95 tok/s 58 tok/s | ~1.8s ~1.8s ~1.8s | |
- | Hy-MT2 1.8B | 1.8B | INT4 Q8 FP16 | 129 tok/s 91 tok/s 56 tok/s | ~1.9s ~1.9s ~1.9s | |
- | CogVideoX 2B | 2B | INT4 Q8 FP16 | 121 tok/s 85 tok/s 51 tok/s | ~2.1s ~2.1s ~2.1s | |
- | LFM3 Audio 2B | 2B | INT4 Q8 FP16 | 121 tok/s 85 tok/s 51 tok/s | ~2.1s ~2.1s ~2.1s | |
- | LFM3 Omni 2B | 2B | INT4 Q8 FP16 | 121 tok/s 85 tok/s 51 tok/s | ~2.1s ~2.1s ~2.1s | |
- | LFM3 VL 2B | 2B | INT4 Q8 FP16 | 121 tok/s 85 tok/s 51 tok/s | ~2.1s ~2.1s ~2.1s | |
- | Liquid D1 2B | 2B | INT4 Q8 FP16 | 121 tok/s 85 tok/s 51 tok/s | ~2.1s ~2.1s ~2.1s | |
- | LTX-Video 0.9.1 | 2B | INT4 Q8 FP16 | 121 tok/s 85 tok/s 51 tok/s | ~2.1s ~2.1s ~2.1s | |
- | Moondream 3.1 | 9B (2B active) | INT4 Q8 FP16 | 121 tok/s 85 tok/s 51 tok/s | ~2.4s ~2.4s ~2.4s | |
- | Stable Diffusion 3 Medium | 2B | INT4 Q8 FP16 | 121 tok/s 85 tok/s 51 tok/s | ~2.1s ~2.1s ~2.1s | |
- | Kandinsky 3.0 | 2.1B | INT4 Q8 FP16 | 118 tok/s 82 tok/s 49 tok/s | ~2.2s ~2.2s ~2.2s | |
- | LTX-2 | 2.2B | INT4 Q8 FP16 | 115 tok/s 79 tok/s 47 tok/s | ~2.3s ~2.3s ~2.3s | |
- | SDXL 1.0 Refiner | 2.31B | INT4 Q8 FP16 | 112 tok/s 77 tok/s 45 tok/s | ~2.4s ~2.4s ~2.4s | |
- | Stable Diffusion 3.5 Medium | 2.5B | INT4 Q8 FP16 | 107 tok/s 72 tok/s 42 tok/s | ~2.6s ~2.6s ~2.6s | |
- | SDXL 1.0 Base | 2.56B | INT4 Q8 FP16 | 105 tok/s 71 tok/s 42 tok/s | ~2.6s ~2.6s ~2.6s | |
- | SDXL Turbo | 2.56B | INT4 Q8 FP16 | 105 tok/s 71 tok/s 42 tok/s | ~2.6s ~2.6s ~2.6s | |
- | Falcon-3B | 3B | INT4 Q8 FP16 | 101 tok/s 66 tok/s 37 tok/s | ~3.0s ~3.0s ~3.0s | |
- | CroissantLLM Base | 1.3B | INT4 Q8 FP16 | 99 tok/s 83 tok/s 58 tok/s | ~1.4s ~1.4s ~1.4s | |
- | Phi-1 | 1.3B | INT4 Q8 FP16 | 99 tok/s 83 tok/s 58 tok/s | ~1.4s ~1.4s ~1.4s | |
- | Phi-1.5 | 1.3B | INT4 Q8 FP16 | 99 tok/s 83 tok/s 58 tok/s | ~1.4s ~1.4s ~1.4s | |
- | DeepSeek-R1 3B | 3B | INT4 Q8 FP16 | 97 tok/s 64 tok/s 36 tok/s | ~3.0s ~3.0s ~3.0s | |
- | Mellum 2.1 12B A2.5B Thinking | 12B (2.5B active) | INT4 Q8 FP16 | 97 tok/s 68 tok/s 39 tok/s | ~3.0s ~3.0s ~3.2s | |
- | Falcon3-3B | 3B | INT4 Q8 FP16 | 95 tok/s 63 tok/s 36 tok/s | ~3.0s ~3.0s ~3.0s | |
- | LFM2-VL 3B | 3B | INT4 Q8 FP16 | 95 tok/s 63 tok/s 36 tok/s | ~3.0s ~3.0s ~3.0s | |
- | Liquid D1 3B | 3B | INT4 Q8 FP16 | 95 tok/s 63 tok/s 36 tok/s | ~3.0s ~3.0s ~3.0s | |
- | Liquid D1-VL 3B | 3B | INT4 Q8 FP16 | 95 tok/s 63 tok/s 36 tok/s | ~3.0s ~3.0s ~3.0s | |
- | Liquid LFM2-VL 3B D1 | 3B | INT4 Q8 FP16 | 95 tok/s 63 tok/s 36 tok/s | ~3.0s ~3.0s ~3.0s | |
- | Qwen2.5-3B | 3B | INT4 Q8 FP16 | 95 tok/s 63 tok/s 36 tok/s | ~3.0s ~3.0s ~3.0s | |
- | LFM2.5 24B A2B | 24B (2.5B active) | INT4 Q8 | 89 tok/s 58 tok/s | ~3.4s ~4.0s | |
- | ERNIE-4.5-21B-A3B | 21B (3B active) | INT4 Q8 | 83 tok/s 55 tok/s | ~3.7s ~4.0s | |
- | ERNIE-4.5-21B-A3B-Base | 21B (3B active) | INT4 Q8 | 83 tok/s 55 tok/s | ~3.7s ~4.0s | |
- | ERNIE-4.5-VL-28B-A3B | 28B (3B active) | INT4 | 80 tok/s | ~3.9s | |
- | ERNIE-4.5-VL-28B-A3B-Base | 28B (3B active) | INT4 | 80 tok/s | ~3.9s | |
- | FLUX.2 Klein 4B | 4B | INT4 Q8 FP16 | 79 tok/s 51 tok/s 28 tok/s | ~4.0s ~4.0s ~4.0s | |
- | FrogNano 4B | 4B | INT4 Q8 FP16 | 79 tok/s 51 tok/s 28 tok/s | ~4.0s ~4.0s ~4.0s | |
- | Hy-Memory 30B A3B | 30B (3B active) | INT4 | 79 tok/s | ~4.0s | |
- | Hy-MT2 30B A3B | 30B (3B active) | INT4 | 79 tok/s | ~4.0s | |
- | Mage-VL Cowork | 4B | INT4 Q8 FP16 | 79 tok/s 51 tok/s 28 tok/s | ~4.0s ~4.0s ~4.0s | |
- | Spark X2.5 4B | 4B | INT4 Q8 FP16 | 79 tok/s 51 tok/s 28 tok/s | ~4.0s ~4.0s ~4.0s | |
- | MiniCPM-V 4.7 35B A3B | 35B (3B active) | INT4 | 78 tok/s | ~4.0s | |
- | EmbeddingGemma 2 | 2B | INT4 Q8 FP16 | 75 tok/s 61 tok/s 41 tok/s | ~2.1s ~2.1s ~2.1s | |
- | CogVideoX 1.5 5B | 5B | INT4 Q8 FP16 | 67 tok/s 42 tok/s 23 tok/s | ~4.9s ~4.9s ~4.9s | |
- | CogVideoX 5B | 5B | INT4 Q8 FP16 | 67 tok/s 42 tok/s 23 tok/s | ~4.9s ~4.9s ~4.9s | |
- | Wan 2.2 TI2V 5B | 5B | INT4 Q8 FP16 | 67 tok/s 42 tok/s 23 tok/s | ~4.9s ~4.9s ~4.9s | |
- | Phi-2 | 2.7B | INT4 Q8 FP16 | 66 tok/s 51 tok/s 33 tok/s | ~2.8s ~2.8s ~2.8s | |
- | MaLLaM-3B | 3B | INT4 Q8 FP16 | 63 tok/s 48 tok/s 31 tok/s | ~3.0s ~3.0s ~3.1s | |
- | SmolLM3 3B | 3B | INT4 Q8 FP16 | 63 tok/s 48 tok/s 31 tok/s | ~3.0s ~3.0s ~3.1s | |
- | Falcon-7B | 7B | INT4 Q8 FP16 | 54 tok/s 32 tok/s 17 tok/s | ~6.8s ~6.8s ~6.8s | |
- | AuraFlow v0.2 | 6.8B | INT4 Q8 FP16 | 53 tok/s 32 tok/s 17 tok/s | ~6.6s ~6.6s ~6.6s | |
- | AuraFlow v0.3 | 6.8B | INT4 Q8 FP16 | 53 tok/s 32 tok/s 17 tok/s | ~6.6s ~6.6s ~6.6s | |
- | Falcon3-7B | 7B | INT4 Q8 FP16 | 52 tok/s 31 tok/s 17 tok/s | ~6.8s ~6.8s ~6.8s | |
- | Hy-MT2 7B | 7B | INT4 Q8 FP16 | 52 tok/s 31 tok/s 17 tok/s | ~6.8s ~6.8s ~6.8s | |
- | Mistral-7B-v0.1 | 7.3B | INT4 Q8 FP16 | 50 tok/s 30 tok/s 16 tok/s | ~7.0s ~7.1s ~7.1s | |
- | Youtu Parsing Omni | 8B | INT4 Q8 FP16 | 48 tok/s 28 tok/s 15 tok/s | ~7.7s ~7.7s ~7.7s | |
- | Stable Diffusion 3.5 Large | 8.1B | INT4 Q8 FP16 | 46 tok/s 28 tok/s 15 tok/s | ~7.8s ~7.8s ~7.8s | |
- | Stable Diffusion 3.5 Large Turbo | 8.1B | INT4 Q8 FP16 | 46 tok/s 28 tok/s 15 tok/s | ~7.8s ~7.8s ~7.8s | |
- | FLUX.2 Klein 9B | 9B | INT4 Q8 FP16 | 42 tok/s 25 tok/s 13 tok/s | ~8.7s ~8.7s ~8.7s | |
- | LensVLM 9B | 9B | INT4 Q8 FP16 | 42 tok/s 25 tok/s 13 tok/s | ~8.7s ~8.7s ~8.7s | |
- | MiMo V2.6 Distill Qwen 9B | 9B | INT4 Q8 FP16 | 42 tok/s 25 tok/s 13 tok/s | ~8.7s ~8.7s ~8.7s | |
- | ChatGLM3-6B-32K | 6B | INT4 Q8 FP16 | 41 tok/s 29 tok/s 17 tok/s | ~5.8s ~5.9s ~5.9s | |
- | Yi-6B | 6B | INT4 Q8 FP16 | 41 tok/s 29 tok/s 17 tok/s | ~5.8s ~5.9s ~5.9s | |
- | Falcon3-10B | 10B | INT4 Q8 FP16 | 38 tok/s 23 tok/s 12 tok/s | ~9.6s ~9.6s ~9.6s | |
- | Kimi-VL-A3B-Instruct | 16B (3B active) | INT4 Q8 | 38 tok/s 33 tok/s | ~3.6s ~3.6s | |
- | Mochi 1 Preview | 10B | INT4 Q8 FP16 | 38 tok/s 23 tok/s 12 tok/s | ~9.6s ~9.6s ~9.6s | |
- | Falcon2-11B | 11B | INT4 Q8 FP16 | 37 tok/s 21 tok/s 11 tok/s | ~10.5s ~10.5s ~11.5s | |
- | Hunyuan Lite | 7B | INT4 Q8 FP16 | 37 tok/s 26 tok/s 15 tok/s | ~6.8s ~6.8s ~6.8s | |
- | OLMo 3 7B Base | 7B | INT4 Q8 FP16 | 37 tok/s 26 tok/s 15 tok/s | ~6.8s ~6.8s ~6.8s | |
- | SEA-LION-7B | 7.1B | INT4 Q8 FP16 | 36 tok/s 25 tok/s 15 tok/s | ~6.9s ~6.9s ~6.9s | |
- | SEA-LION-7B-Instruct | 7.1B | INT4 Q8 FP16 | 36 tok/s 25 tok/s 15 tok/s | ~6.9s ~6.9s ~6.9s | |
- | FLUX.1 Canny Dev | 11.9B | INT4 Q8 FP16 | 33 tok/s 19 tok/s 10 tok/s | ~11.4s ~11.4s ~12.4s | |
- | FLUX.1 Depth Dev | 11.9B | INT4 Q8 FP16 | 33 tok/s 19 tok/s 10 tok/s | ~11.4s ~11.4s ~12.4s | |
- | FLUX.1 Dev | 11.9B | INT4 Q8 FP16 | 33 tok/s 19 tok/s 10 tok/s | ~11.4s ~11.4s ~12.4s | |
- | FLUX.1 Fill Dev | 11.9B | INT4 Q8 FP16 | 33 tok/s 19 tok/s 10 tok/s | ~11.4s ~11.4s ~12.4s | |
- | FLUX.1 Schnell | 11.9B | INT4 Q8 FP16 | 33 tok/s 19 tok/s 10 tok/s | ~11.4s ~11.4s ~12.4s | |
- | Sahabat-AI-Llama3-8B-Instruct | 8B | INT4 Q8 FP16 | 33 tok/s 23 tok/s 13 tok/s | ~7.7s ~7.7s ~7.8s | |
- | Typhoon-2-8B | 8B | INT4 Q8 FP16 | 33 tok/s 23 tok/s 13 tok/s | ~7.7s ~7.7s ~7.8s | |
- | GLM-4-9B | 9B | INT4 Q8 FP16 | 30 tok/s 21 tok/s 12 tok/s | ~8.7s ~8.7s ~8.7s | |
- | GLM-4-9B-Chat | 9B | INT4 Q8 FP16 | 30 tok/s 21 tok/s 12 tok/s | ~8.7s ~8.7s ~8.7s | |
- | GLM-4-9B-Chat-1M | 9B | INT4 Q8 FP16 | 30 tok/s 21 tok/s 12 tok/s | ~8.7s ~8.7s ~8.7s | |
- | HunyuanVideo 1.5 720P T2V | 13B | INT4 Q8 FP16 | 30 tok/s 18 tok/s 8 tok/s | ~12.4s ~12.4s ~14.7s | |
- | Sahabat-AI-Gemma2-9B | 9.2B | INT4 Q8 FP16 | 30 tok/s 20 tok/s 11 tok/s | ~8.8s ~8.9s ~9.7s | |
- | Sahabat-AI-Gemma2-9B-Instruct | 9.2B | INT4 Q8 FP16 | 30 tok/s 20 tok/s 11 tok/s | ~8.8s ~8.9s ~9.7s | |
- | Yi-9B | 9B | INT4 Q8 FP16 | 30 tok/s 21 tok/s 12 tok/s | ~8.7s ~8.7s ~8.7s | |
- | Wan 2.1 I2V 14B 720P | 14B | INT4 Q8 | 29 tok/s 17 tok/s | ~13.4s ~13.4s | |
- | Wan 2.1 T2V 14B | 14B | INT4 Q8 | 29 tok/s 17 tok/s | ~13.4s ~13.4s | |
- | Wan 2.2 T2V A14B | 14B | INT4 Q8 | 29 tok/s 17 tok/s | ~13.4s ~13.4s | |
- | FLUX.2 Dev | 15B | INT4 Q8 | 27 tok/s 16 tok/s | ~14.3s ~14.3s | |
- | GreenMind-14B-R1 | 14B | INT4 Q8 | 21 tok/s 14 tok/s | ~13.4s ~13.4s | |
- | Arete 27B | 27B | INT4 | 16 tok/s | ~25.4s | |
- | AREX-2 | 27B | INT4 | 16 tok/s | ~25.4s | |
- | PPLX Decider v1 27B | 27B | INT4 | 16 tok/s | ~25.4s | |
- | PPLX Decider v1.1 27B | 27B | INT4 | 16 tok/s | ~25.4s | |
- | Agnes 3.0 Flash | 33B | INT4 | 13 tok/s | ~30.9s | |
- | Falcon-40B | 40B | INT4 | 11 tok/s | ~40.6s | |
Inference Capacity
Maximum model parameter class runnable by weight precision.
Context:
4-bit
32B Models
8-bit
14B Models
16-bit
8B Models
Fine-Tuning Capacity
Hardware capacity for local training and adapter fine-tuning.
Context Length:
QLoRA
32B Models
LoRA
14B Models
Full Parameter Training
3B Models
Workload Recommendations
Guidance for optimal precision and operational ceilings on Apple M3 Max (36GB).
Inference Sweet Spot
Optimized for 14B models at uncompressed or Q8 precision, and 32B models at Q4 with up to 32k context.
Bandwidth & Speed Profile
With 300 GB/s aggregate bandwidth, batch size 1 inference operates in a memory-bandwidth bound regime. Generates approximately 67 tok/s on an 8B Q4 model and 8 tok/s on a 70B Q4 model.
Unified Memory Budget
Modeled with a 75% usable allocation budget to preserve system RAM for macOS and display compositor processes.
Frequently Asked Questions
©2025 ApX Machine Learning