M5 Pro 48GB vs AMD Strix Halo 128GB for local LLMs
9 reports on the M5 Pro 48GB and 318 on the AMD Strix Halo 128GB
Which is faster for local LLMs?
On Qwen3.5 27B at 4-bit, the M5 Pro 48GB has generation speed not reported; prompt processing 461.0 tokens/s (one run); the AMD Strix Halo 128GB has 12.1 generation tokens/s (one run); prompt processing 310.0 tokens/s (one run). Memory bandwidth, which caps how fast a card can write, is 20% higher on the M5 Pro 48GB (307 GB/s on the M5 Pro 48GB, 256 on the AMD Strix Halo 128GB).
- Memory
- 48GB unified
- Memory bandwidth
- 307 GB/s
- FP16 compute
- not on record
- Reports
- 9
- Memory
- 128GB unified
- Memory bandwidth
- 256 GB/s
- FP16 compute
- 59 TFLOPS
- Reports
- 318
Measured on both
Generation and prompt processing speeds from plain runs (one device, one request, no speculative decoding, the whole model in the device's memory) of the same model size at the same bit class. Each side shows the engines and reported context behind its median. Context can mean a configured window or prompt depth, so these are community comparisons with differing settings. How to read these
| Model | M5 Pro 48GB | AMD Strix Halo 128GB |
|---|---|---|
| Qwen3.5 27B4-bit | -0 runsPP 461.0 (one run)engine not stated4K context | 12.1one runPP 310.0 (one run)engine not statedcontext not stated |
Estimated by the calculator
The quant the calculator recommends for each card and its estimated speed at a 32,768-token context with 32 GB of system RAM, from memory bandwidth and the model's shape. Where a model was also measured above, the measurement wins.
| Model | M5 Pro 48GB | AMD Strix Halo 128GB |
|---|---|---|
| Qwen3.5 27B | 12.0 t/sQ6_K | 3.5 t/sFP16 |
| Qwen3.8 27B | 12.0 t/sQ6_K | 3.5 t/sFP16 |
| Qwen3.8 125B · 6B active | doesn't fit at these settings | 12.6 t/sQ6_K |
| Qwen3.6 35B · 3B active | 47.5 t/sQ5_K_M | 12.6 t/sFP16 |
| DeepSeek V4 Flash 284B · 13B active | doesn't fit at these settings | 33.5 t/sQ2_K |
| Qwen3.6 27B | 12.0 t/sQ6_K | 3.5 t/sFP16 |
| Gemma 4 26B · 4B active | 23.2 t/sQ8_0 | 9.5 t/sFP16 |
Change the context, RAM or model in the calculator: M5 Pro 48GB · AMD Strix Halo 128GB
Every report on each card, with its source: M5 Pro 48GB · AMD Strix Halo 128GB