← All hardware
Apple's mainstream silicon that puts a neural accelerator inside every GPU core for a big on-device AI leap.
Pros
- Massive generational jump in GPU/AI compute (~3.5x M4)
- Excellent performance-per-watt and battery life
- Unified memory makes mid-size local models practical
Cons
- Base 16GB/32GB memory ceiling limits large local LLMs
- macOS/Apple ecosystem lock-in; no CUDA
- Sold only inside Apple devices, not as a standalone chip
✓ Where it shines / best for
- On-device AI and local model inference on Apple silicon
- Creators needing strong GPU compute for AI tasks
- Mainstream pro Mac and iPad Pro buyers
✕ Not the best fit for
- Highest-end pro video/3D needing Pro/Max variants
- Non-Apple ecosystems
- Standalone retail purchase
Features
- ✓ Up to 10-core CPU
- ✓ Up to 10-core GPU with a Neural Accelerator built into each core
- ✓ Over 4x peak GPU compute for AI compared to M4
- ✓ 16-core Neural Engine improved over M4
- ✓ Up to 153 GB/s unified memory bandwidth (LPDDR5X 9600 MT/s)
- ✓ Up to 32GB unified memory
- ✓ Third-generation 3nm process
- ✓ Hardware ray tracing and Dynamic Caching, optimized for on-device AI
Pricing
| Plan | Price | Billing | Notes |
|---|---|---|---|
| Embedded in Mac / iPad / Vision Pro | $1,599 | one-time | Same silicon as the M5; not sold separately. Ships in 14" MacBook Pro (from $1,599), iPad Pro (from $999), and Apple Vision Pro (from $3,499). |
Pricing verified from the official source. Prices change often — confirm on the vendor's site before buying.
Specifications
| memory | Up to 32GB LPDDR5X unified |
| cpu_cores | 10 (4P + 6E) |
| gpu_cores | 10 with per-core Neural Accelerator |
| architecture | Apple M5, ~3nm TSMC, 10-core CPU + 10-core GPU |
| neural_engine | 16-core |
| ai_performance | ~3.5x M4 GPU AI compute (Apple claim) |
| memory_bandwidth | 153 GB/s |
Sponsored
A full review is being generated for this product and will appear here shortly.