Hardware · Apple
What AI models can the Apple iPhone 12 Pro / 12 Pro Max run?
- RAM
- 6.0 GB
- Storage
- 128 GB – 512 GB
- One app can use
- 3.0 GB
- Memory bandwidth
- 34 GB/s
- FP16 compute
- 1.0 TFLOPS
- Released
- 2020
Every iPhone 12 Pro / 12 Pro Max has 6 GB of RAM, whatever its storage (128 GB, 256 GB or 512 GB). A model runs in RAM (the 6.0 GB here), not in storage. Storage (128 GB, 256 GB or 512 GB on this version) only keeps the downloaded files, and even the smallest size has room for several. iOS lets one app hold only part of the RAM before it closes the app, so we plan around 3.0 GB; a model bigger than that won’t run. Phone apps such as PocketPal AI download a model once, then run it with no connection.
More about this hardware
| Maker | Apple |
|---|---|
| Type | Mobile device (system on a chip) |
| App share of memory | 50% (what one app can typically hold before the system closes it) |
| Chip | Apple A14 |
| RAM type | 6 GB LPDDR4X-4266 (teardown) |
| Versions sold | 12 Pro: 6 GB RAM with 128 GB, 256 GB or 512 GB 12 Pro Max: 6 GB RAM with 128 GB, 256 GB or 512 GB |
| Bandwidth figure | 34.1 GB/s, from teardowns (the maker publishes no memory speed) |
| Note | A14 with a 4-core GPU; 6 GB LPDDR4X-4266 (34.1 GB/s). |
| Maker’s specifications | www.apple.com |
Apps that run models well on the Apple iPhone 12 Pro / 12 Pro Max
iOS
Best LLMs for the Apple iPhone 12 Pro / 12 Pro Max
| Quality | Model | Quant | Memory | Speed | Runs |
|---|---|---|---|---|---|
| 57Good quality (B) · 57 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | MiniCPM5 2B | Q4_K_M | 2.5 GB | 17.1 tok/s | Runs well |
| 53Fair quality (C) · 53 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | LFM3 1.7B | Q4_K_M | 1.8 GB | 24.9 tok/s | Runs well |
| 55Good quality (B) · 55 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | Spark-X2.6 4B | Q3_K_M | 2.9 GB | 13.5 tok/s | Runs welltight fit |
| 47Fair quality (C) · 47 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | LFM3-VL 2B | Q4_K_M | 1.9 GB | 22.4 tok/s | Runs well |
| 43Fair quality (C) · 43 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. | Gemma 4 E2B | Q2_K | 2.7 GB | 30.7 tok/s | Runs welltight fit |
| 51Fair quality (C) · 51 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | TMax 4B | Q3_K_M | 2.9 GB | 13.2 tok/s | Runs welltight fit |
| 36Basic quality (D) · 36 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | Unlimited OCR 2 (3B-A0.6B) | Q3_K_M | 2.7 GB | 71.3 tok/s | Runs welltight fit |
| 47Fair quality (C) · 47 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | Nemotron Labs Dory 1.0 4B | Q2_K | 2.7 GB | 17.8 tok/s | Runs welltight fit |
| 38Basic quality (D) · 38 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | TeleOCR | Q4_K_M | 2.4 GB | 30.3 tok/s | Runs great |
| 33Basic quality (D) · 33 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | Unlimited OCR (3B-A0.6B) | Q3_K_M | 2.7 GB | 71.3 tok/s | Runs welltight fit |
| 41Fair quality (C) · 41 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | LFM2.5 2.6B | Q4_K_M | 2.4 GB | 15.9 tok/s | Runs well |
| 39Basic quality (D) · 39 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. | PaddleOCR-VL 1.6 0.9B | Q4_K_M | 1.3 GB | 44.8 tok/s | Runs great |
Best Image models for the Apple iPhone 12 Pro / 12 Pro Max
| Quality | Model | Quant | Memory | Speed | Runs |
|---|---|---|---|---|---|
| 30Basic quality (D) · 30 of 100: how capable the model is at FP16, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. | LCM DreamShaper v7 | FP16 | 2.8 GB | 13 s | Runs welltight fit |
| 42Fair quality (C) · 42 of 100: how capable the model is at FP16, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | Realistic Vision V5.1 | FP16 | 2.8 GB | 62 s | Slowtight fit |
| 40Fair quality (C) · 40 of 100: how capable the model is at FP16, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking. | DreamShaper 8 | FP16 | 2.8 GB | 62 s | Slowtight fit |
| 28Basic quality (D) · 28 of 100: how capable the model is at FP16, whatever it runs on. Match score 60/100 (quality, fit, speed and memory use) sets the ranking. | SD Turbo | FP16 | 2.6 GB | 4.4 s | Runs welltight fit · encoder unloaded after use |
| 35Basic quality (D) · 35 of 100: how capable the model is at FP16, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking. | Stable Diffusion 1.5 | FP16 | 2.8 GB | 62 s | Slowtight fit |
Best Video models for the Apple iPhone 12 Pro / 12 Pro Max
None of the video models in our catalog fit in this much memory.
Best Decision models for the Apple iPhone 12 Pro / 12 Pro Max
None of the decision models in our catalog fit in this much memory.
Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.