iPad Pro M4 16GB

iOS High tier · Apple M4 · 16 GB RAM · 120 GB/s bandwidth
View device page →
BENCHMARK RESULTS (69)
Model Quant Measured Estimated RAM used App Source Date
🗣️ KittenTTS Mini FP16 224 tok/s 200 tok/s +12% 0.2 GB LocallyAI community 2026-05-13
🔢 BGE Small FP16 222 tok/s 200 tok/s +11% 0.1 GB LocallyAI community 2026-03-15
🔢 Nomic Embed Text FP16 215 tok/s 200 tok/s +8% 0.3 GB LocallyAI editorial 2026-03-05
SmolLM2 135M SmolLM2 135M FP16 207 tok/s 200 tok/s +4% 0.3 GB LocallyAI community 2026-04-06
Ternary Bonsai 1.7B Ternary Bonsai 1.7B Q2 197 tok/s 200 tok/s -1% 0.5 GB LocallyAI community 2026-03-16
🗣️ Kokoro 82M FP16 192 tok/s 200 tok/s -4% 0.5 GB LocallyAI community 2026-03-26
🔢 all-MiniLM-L6-v2 FP16 186 tok/s 200 tok/s -7% 0.1 GB LocallyAI community 2026-04-10
Whisper Tiny Whisper Tiny FP16 185 tok/s 200 tok/s -7% 0.1 GB Whisper Transcription community 2026-03-27
🗣️ KittenTTS Nano FP16 182 tok/s 200 tok/s -9% 0.1 GB LocallyAI community 2026-05-14
Whisper Small Whisper Small FP16 178 tok/s 171 tok/s +4% 0.6 GB Whisper Transcription community 2026-02-07
Wav2Vec2 Base Wav2Vec2 Base FP16 176 tok/s 200 tok/s -12% 0.3 GB Whisper Transcription community 2026-05-13
🔢 GTE Large FP16 117 tok/s 125 tok/s -6% 0.8 GB LocallyAI community 2026-05-11
Ternary Bonsai 4B Ternary Bonsai 4B Q2 95 tok/s 98 tok/s -3% 1.1 GB LocallyAI community 2026-04-04
SmolLM2 360M SmolLM2 360M FP16 91 tok/s 91 tok/s +0% 1.1 GB LocallyAI community 2026-05-24
🗣️ OuteTTS 0.3 500M FP16 87 tok/s 84 tok/s +4% 1.2 GB LocallyAI community 2026-01-16
👁️ SmolVLM 500M FP16 80 tok/s 76 tok/s +5% 1.4 GB LocallyAI community 2026-05-28
Qwen 2.5 0.5B Qwen 2.5 0.5B FP16 78 tok/s 70 tok/s +11% 1.3 GB LocallyAI community 2026-05-12
Bonsai 8B (1-bit) Bonsai 8B (1-bit) Q1 78 tok/s 72 tok/s +8% 1.3 GB LocallyAI community 2026-01-03
Qwen 2.5 Coder 0.5B Qwen 2.5 Coder 0.5B FP16 75 tok/s 70 tok/s +7% 1.4 GB LocallyAI community 2026-04-19
Gemma 3 1B Gemma 3 1B Q8 62 tok/s 38 tok/s +63% 1.5 GB LM Studio editorial 2026-03-10
🗣️ NeuTTS Air FP16 62 tok/s 56 tok/s +11% 1.8 GB LocallyAI community 2026-03-17
🎙️ Distil-Whisper Large V3 FP16 62 tok/s 56 tok/s +11% 1.8 GB Whisper Transcription community 2026-04-20
Bonsai Image 4B Bonsai Image 4B Q2 52 tok/s 47 tok/s +11% 2.0 GB LocallyAI editorial 2026-02-11
Qwen 3.5 0.8B Qwen 3.5 0.8B FP16 51 tok/s 47 tok/s +9% 1.9 GB LocallyAI community 2026-03-04
Whisper Medium Whisper Medium FP16 51 tok/s 56 tok/s -9% 1.9 GB Whisper Transcription editorial 2026-04-19
Ternary Bonsai 8B Ternary Bonsai 8B Q2 50 tok/s 53 tok/s -6% 2.0 GB LocallyAI community 2026-02-24
TinyLlama 1.1B TinyLlama 1.1B FP16 41 tok/s 38 tok/s +8% 2.4 GB LocallyAI community 2026-02-27
Phi-4 Mini Phi-4 Mini Q4 40 tok/s 19 tok/s +111% 2.4 GB LM Studio editorial 2026-03-10
Gemma 3 1B Gemma 3 1B FP16 39 tok/s 38 tok/s +3% 2.8 GB LocallyAI editorial 2026-02-25
LFM2.5 1.2B LFM2.5 1.2B FP16 37 tok/s 36 tok/s +3% 3.0 GB LocallyAI editorial 2026-03-04
🎨 Stable Diffusion Turbo FP16 37 tok/s 40 tok/s -7% 2.4 GB LocallyAI community 2026-04-09
Whisper Large V3 Whisper Large V3 FP16 29 tok/s 27 tok/s +7% 3.5 GB Whisper Transcription community 2026-03-12
Llama 3.2 1B Llama 3.2 1B FP16 28 tok/s 31 tok/s -10% 3.5 GB LocallyAI editorial 2026-02-10
🧠 DeepSeek R1 Distill 1.5B FP16 28 tok/s 25 tok/s +12% 3.7 GB LocallyAI editorial 2026-01-03
SmolLM2 1.7B SmolLM2 1.7B FP16 27 tok/s 25 tok/s +8% 3.7 GB LocallyAI community 2026-05-26
Qwen 2.5 Coder 1.5B Qwen 2.5 Coder 1.5B FP16 26 tok/s 26 tok/s +0% 3.6 GB LocallyAI editorial 2026-05-08
StableLM 2 1.6B StableLM 2 1.6B FP16 25 tok/s 25 tok/s +0% 4.2 GB LocallyAI editorial 2026-02-25
👁️ Moondream 2B FP16 24 tok/s 23 tok/s +4% 4.1 GB LocallyAI editorial 2026-05-12
🗣️ Dia 1.6B FP16 22 tok/s 25 tok/s -12% 4.1 GB LocallyAI editorial 2026-04-22
Qwen 3.5 4B Qwen 3.5 4B Q8 19 tok/s 19 tok/s +0% 4.8 GB LocallyAI community 2026-05-28
Gemma 3 4B Gemma 3 4B Q8 18 tok/s 17 tok/s +6% 5.2 GB LocallyAI community 2026-05-25
Gemma 4 E2B Gemma 4 E2B FP16 18 tok/s 19 tok/s -5% 5.8 GB LocallyAI community 2026-03-28
Qwen 3.5 2B Qwen 3.5 2B FP16 18 tok/s 20 tok/s -10% 5.2 GB LocallyAI editorial 2026-01-01
Phi-4 Mini Phi-4 Mini Q8 18 tok/s 19 tok/s -5% 5.7 GB LocallyAI community 2026-04-06
Gemma 3n E2B Gemma 3n E2B Q8 17 tok/s 18 tok/s -6% 6.1 GB LocallyAI editorial 2026-05-27
🎨 Stable Diffusion 3.5 Medium FP16 17 tok/s 18 tok/s -6% 5.5 GB LocallyAI community 2026-02-10
🧠 DeepSeek R1 Distill 7B Q6 16 tok/s 14 tok/s +14% 7.1 GB LocallyAI community 2026-03-16
Mistral 7B Mistral 7B Q6 16 tok/s 14 tok/s +14% 6.7 GB LocallyAI community 2026-03-04
👁️ LLaVA 1.6 7B Q6 16 tok/s 15 tok/s +7% 6.4 GB LocallyAI community 2026-03-14
Qwen 2.5 7B Qwen 2.5 7B Q6 15 tok/s 14 tok/s +7% 7.6 GB LocallyAI editorial 2026-03-02
Gemma 4 E4B Gemma 4 E4B Q8 15 tok/s 17 tok/s -12% 6.3 GB LocallyAI community 2026-02-24
Qwen 2.5 VL 7B Qwen 2.5 VL 7B Q6 15 tok/s 14 tok/s +7% 6.8 GB LocallyAI community 2026-01-04
Gemma 3n E4B Gemma 3n E4B Q6 14 tok/s 13 tok/s +8% 7.7 GB LocallyAI community 2026-04-19
LFM2.5 8B A1B LFM2.5 8B A1B Q6 14 tok/s 13 tok/s +8% 7.7 GB LocallyAI community 2026-03-14
Qwen 2.5 Coder 7B Qwen 2.5 Coder 7B Q6 14 tok/s 14 tok/s +0% 6.5 GB LocallyAI community 2026-02-10
🎨 SDXL Turbo FP16 14 tok/s 13 tok/s +8% 7.2 GB LocallyAI community 2026-04-06
Llama 3.2 3B Llama 3.2 3B FP16 13 tok/s 13 tok/s +0% 7.4 GB LocallyAI community 2026-03-12
Gemma 3 12B Gemma 3 12B Q4 13 tok/s 12 tok/s +8% 8.9 GB LocallyAI editorial 2026-02-27
Mistral Nemo 12B Mistral Nemo 12B Q4 13 tok/s 12 tok/s +8% 8.0 GB LocallyAI community 2026-04-19
🧠 DeepSeek R1 Distill 8B Q6 13 tok/s 14 tok/s -7% 7.4 GB LocallyAI community 2026-03-13
👨‍💻 DeepSeek Coder 6.7B Q8 13 tok/s 12 tok/s +8% 9.2 GB LocallyAI editorial 2026-03-01
Qwen 2.5 3B Qwen 2.5 3B FP16 12 tok/s 13 tok/s -8% 7.4 GB LocallyAI community 2026-04-10
Llama 3.1 8B Llama 3.1 8B Q6 12 tok/s 14 tok/s -14% 7.5 GB LocallyAI community 2026-05-23
🧠 Qwen3 8B Q6 12 tok/s 13 tok/s -8% 7.5 GB LocallyAI community 2026-04-19
Phi-4 Medium Phi-4 Medium Q3 12 tok/s 14 tok/s -14% 6.7 GB LocallyAI editorial 2026-02-08
👨‍💻 StarCoder2 3B FP16 12 tok/s 14 tok/s -14% 7.6 GB LocallyAI community 2026-01-17
🎨 FLUX.1 Schnell Q4 12 tok/s 12 tok/s +0% 8.8 GB LocallyAI editorial 2026-05-25
Qwen 3.5 9B Qwen 3.5 9B Q6 11 tok/s 12 tok/s -8% 8.8 GB LocallyAI community 2026-02-22
Qwen 2.5 Coder 3B Qwen 2.5 Coder 3B FP16 11 tok/s 13 tok/s -15% 7.7 GB LocallyAI community 2026-02-16
← All benchmarks