MacBook Pro M3 Pro 18GB

macOS High tier · Apple M3 Pro · 18 GB RAM · 150 GB/s bandwidth
View device page →
BENCHMARK RESULTS (73)
Model Quant Measured Estimated RAM used App Source Date
Whisper Small Whisper Small FP16 222 tok/s 200 tok/s +11% 0.6 GB MacWhisper community 2026-01-18
Ternary Bonsai 1.7B Ternary Bonsai 1.7B Q2 219 tok/s 200 tok/s +10% 0.4 GB Ollama community 2026-03-04
🔢 all-MiniLM-L6-v2 FP16 214 tok/s 200 tok/s +7% 0.1 GB Xybrid CLI community 2026-05-26
🔢 GTE Large FP16 203 tok/s 200 tok/s +2% 0.8 GB LM Studio editorial 2026-05-12
🗣️ KittenTTS Nano FP16 200 tok/s 200 tok/s +0% 0.1 GB Xybrid CLI community 2026-04-21
SmolLM2 135M SmolLM2 135M FP16 199 tok/s 200 tok/s +0% 0.3 GB Xybrid CLI editorial 2026-01-17
Wav2Vec2 Base Wav2Vec2 Base FP16 199 tok/s 200 tok/s +0% 0.2 GB MacWhisper community 2026-05-10
🔢 BGE Small FP16 183 tok/s 200 tok/s -8% 0.1 GB Ollama community 2026-03-17
🗣️ KittenTTS Mini FP16 180 tok/s 200 tok/s -10% 0.2 GB Piper community 2026-02-09
🔢 Nomic Embed Text FP16 179 tok/s 200 tok/s -10% 0.3 GB Jan community 2026-03-24
Whisper Tiny Whisper Tiny FP16 179 tok/s 200 tok/s -10% 0.1 GB Xybrid CLI community 2026-05-12
🗣️ Kokoro 82M FP16 178 tok/s 200 tok/s -11% 0.4 GB Xybrid CLI community 2026-05-11
Ternary Bonsai 4B Ternary Bonsai 4B Q2 165 tok/s 174 tok/s -5% 1.0 GB Xybrid CLI community 2026-05-15
SmolLM2 360M SmolLM2 360M FP16 158 tok/s 163 tok/s -3% 1.2 GB Ollama community 2026-05-26
🗣️ OuteTTS 0.3 500M FP16 153 tok/s 150 tok/s +2% 1.3 GB Piper editorial 2026-01-21
👁️ SmolVLM 500M FP16 146 tok/s 136 tok/s +7% 1.3 GB Xybrid CLI community 2026-04-20
Qwen 2.5 0.5B Qwen 2.5 0.5B FP16 133 tok/s 125 tok/s +6% 1.3 GB Xybrid CLI community 2026-05-01
Qwen 2.5 Coder 0.5B Qwen 2.5 Coder 0.5B FP16 121 tok/s 125 tok/s -3% 1.3 GB Xybrid CLI community 2026-05-07
Bonsai 8B (1-bit) Bonsai 8B (1-bit) Q1 120 tok/s 129 tok/s -7% 1.3 GB Jan community 2026-04-10
Whisper Medium Whisper Medium FP16 102 tok/s 100 tok/s +2% 1.9 GB Xybrid CLI community 2026-05-26
Gemma 3 1B Gemma 3 1B Q8 98 tok/s 68 tok/s +44% 1.5 GB Ollama editorial 2026-03-20
🎙️ Distil-Whisper Large V3 FP16 94 tok/s 100 tok/s -6% 1.6 GB Xybrid CLI community 2026-03-26
🗣️ NeuTTS Air FP16 90 tok/s 100 tok/s -10% 1.9 GB Xybrid CLI community 2026-03-27
Qwen 3.5 0.8B Qwen 3.5 0.8B FP16 89 tok/s 83 tok/s +7% 2.1 GB LM Studio community 2026-05-10
Ternary Bonsai 8B Ternary Bonsai 8B Q2 87 tok/s 94 tok/s -7% 2.0 GB LM Studio editorial 2026-01-02
Bonsai Image 4B Bonsai Image 4B Q2 81 tok/s 83 tok/s -2% 2.2 GB Xybrid CLI community 2026-05-16
TinyLlama 1.1B TinyLlama 1.1B FP16 73 tok/s 68 tok/s +7% 2.6 GB LM Studio community 2026-05-10
🎨 Stable Diffusion Turbo FP16 73 tok/s 71 tok/s +3% 2.6 GB Jan community 2026-05-27
Gemma 3 1B Gemma 3 1B FP16 70 tok/s 68 tok/s +3% 2.5 GB LM Studio editorial 2026-05-12
LFM2.5 1.2B LFM2.5 1.2B FP16 67 tok/s 64 tok/s +5% 2.8 GB Ollama community 2026-05-11
Phi-4 Mini Phi-4 Mini Q4 62 tok/s 19 tok/s +226% 2.4 GB LM Studio editorial 2026-03-20
Whisper Large V3 Whisper Large V3 FP16 53 tok/s 48 tok/s +10% 3.4 GB MacWhisper community 2026-01-06
Llama 3.2 1B Llama 3.2 1B FP16 52 tok/s 56 tok/s -7% 2.8 GB Xybrid CLI community 2026-03-25
🗣️ Dia 1.6B FP16 48 tok/s 45 tok/s +7% 3.9 GB Piper community 2026-01-16
🧠 DeepSeek R1 Distill 1.5B FP16 47 tok/s 45 tok/s +4% 4.2 GB LM Studio editorial 2026-02-24
StableLM 2 1.6B StableLM 2 1.6B FP16 44 tok/s 45 tok/s -2% 3.7 GB Xybrid CLI editorial 2026-04-06
Qwen 2.5 Coder 1.5B Qwen 2.5 Coder 1.5B FP16 43 tok/s 47 tok/s -9% 3.8 GB Jan community 2026-05-12
SmolLM2 1.7B SmolLM2 1.7B FP16 39 tok/s 44 tok/s -11% 3.8 GB Ollama community 2026-04-11
Qwen 3.5 2B Qwen 3.5 2B FP16 39 tok/s 36 tok/s +8% 4.7 GB Xybrid CLI editorial 2026-05-28
👁️ Moondream 2B FP16 38 tok/s 41 tok/s -7% 3.9 GB LM Studio community 2026-04-22
🎨 Stable Diffusion 3.5 Medium FP16 35 tok/s 33 tok/s +6% 5.3 GB Xybrid CLI community 2026-04-19
Gemma 4 E2B Gemma 4 E2B FP16 34 tok/s 33 tok/s +3% 5.6 GB LM Studio editorial 2026-02-23
Qwen 2.5 3B Qwen 2.5 3B FP16 24 tok/s 23 tok/s +4% 6.9 GB Xybrid CLI community 2026-01-05
👨‍💻 StarCoder2 3B FP16 24 tok/s 24 tok/s +0% 7.8 GB LM Studio editorial 2026-02-26
Qwen 2.5 Coder 3B Qwen 2.5 Coder 3B FP16 24 tok/s 23 tok/s +4% 8.1 GB Xybrid CLI community 2026-05-26
Llama 3.2 3B Llama 3.2 3B FP16 22 tok/s 23 tok/s -4% 7.8 GB Ollama community 2026-03-13
Gemma 3n E4B Gemma 3n E4B Q8 22 tok/s 20 tok/s +10% 8.9 GB Ollama community 2026-02-21
🎨 SDXL Turbo FP16 22 tok/s 23 tok/s -4% 7.8 GB Ollama community 2026-03-15
👨‍💻 DeepSeek Coder 6.7B Q8 21 tok/s 21 tok/s +0% 9.3 GB Jan community 2026-05-24
Llama 3.1 8B Llama 3.1 8B Q8 19 tok/s 18 tok/s +6% 9.1 GB Jan community 2026-05-24
🧠 DeepSeek R1 Distill 8B Q8 19 tok/s 18 tok/s +6% 10.2 GB Xybrid CLI community 2026-04-01
Qwen 2.5 VL 7B Qwen 2.5 VL 7B Q8 19 tok/s 18 tok/s +6% 10.8 GB LM Studio editorial 2026-02-23
Qwen 2.5 7B Qwen 2.5 7B Q8 18 tok/s 19 tok/s -5% 10.5 GB Xybrid CLI community 2026-03-01
Gemma 4 E4B Gemma 4 E4B FP16 18 tok/s 17 tok/s +6% 9.5 GB LM Studio community 2026-03-28
Phi-4 Mini Phi-4 Mini FP16 18 tok/s 19 tok/s -5% 9.1 GB Jan community 2026-05-12
🧠 DeepSeek R1 Distill 7B Q8 18 tok/s 19 tok/s -5% 9.6 GB Ollama community 2026-04-19
👁️ LLaVA 1.6 7B Q8 18 tok/s 20 tok/s -10% 8.6 GB LM Studio editorial 2026-01-05
Gemma 3 4B Gemma 3 4B FP16 17 tok/s 17 tok/s +0% 9.4 GB Ollama community 2026-01-05
Qwen 3.5 4B Qwen 3.5 4B FP16 17 tok/s 18 tok/s -6% 9.9 GB Ollama community 2026-04-21
LFM2.5 8B A1B LFM2.5 8B A1B Q8 17 tok/s 17 tok/s +0% 9.3 GB Xybrid CLI community 2026-04-02
Mistral 7B Mistral 7B Q8 17 tok/s 18 tok/s -6% 9.8 GB Ollama community 2026-03-15
🧠 Qwen3 8B Q8 17 tok/s 17 tok/s +0% 9.9 GB Jan community 2026-01-02
Qwen 2.5 Coder 7B Qwen 2.5 Coder 7B Q8 17 tok/s 19 tok/s -11% 9.9 GB Jan community 2026-03-27
Gemma 3n E2B Gemma 3n E2B FP16 16 tok/s 17 tok/s -6% 11.5 GB Jan community 2026-02-21
Gemma 3 12B Gemma 3 12B Q6 16 tok/s 16 tok/s +0% 10.6 GB LM Studio editorial 2026-05-08
Qwen 3.5 9B Qwen 3.5 9B Q8 16 tok/s 15 tok/s +7% 12.5 GB Ollama editorial 2026-04-06
👨‍💻 Laguna XS.2 Q2 14 tok/s 14 tok/s +0% 12.1 GB LM Studio community 2026-04-04
Gemma 4 31B Gemma 4 31B Q2 13 tok/s 14 tok/s -7% 12.4 GB Ollama community 2026-02-11
Mistral Nemo 12B Mistral Nemo 12B Q8 12 tok/s 12 tok/s +0% 15.7 GB Jan community 2026-05-26
Qwen 3.5 35B A3B Qwen 3.5 35B A3B Q2 12 tok/s 13 tok/s -8% 15.5 GB LM Studio editorial 2026-04-21
Phi-4 Medium Phi-4 Medium Q6 12 tok/s 14 tok/s -14% 13.5 GB Xybrid CLI editorial 2026-04-22
Gemma 4 26B A4B Gemma 4 26B A4B Q3 11 tok/s 12 tok/s -8% 15.0 GB Ollama community 2026-04-20
🎨 FLUX.1 Schnell Q8 11 tok/s 12 tok/s -8% 14.1 GB LM Studio editorial 2026-03-14
← All benchmarks