Home › GPUs & Macs › Apple M2 (10-core GPU) vs Apple M2 (8-core GPU)
Apple M2 (10-core GPU) vs Apple M2 (8-core GPU) for local LLMs
Across 5 tests measured on both, the two are about even (median ratio 1.01×).
On Llama 2 7B (Q4_0, llama-bench) the Apple M2 (10-core GPU) generates 21.9 tokens/s and the Apple M2 (8-core GPU) 21.7 tokens/s (measured).
Same model, same quant, same test
| Model | Quant / test | M2 (10-core GPU) tok/s | M2 (8-core GPU) tok/s | Faster | Sources and run details |
|---|---|---|---|---|---|
| Llama 2 7B | Q4_0 llama-bench | 21.9 Metal | 21.7 Metal | even | llama.cpp #4167 M2 (10-core GPU) · llama.cpp #4167 M2 (8-core GPU) |
| Llama 2 7B | Q8_0 llama-bench | 12.2 Metal | 12.2 Metal | even | llama.cpp #4167 M2 (10-core GPU) · llama.cpp #4167 M2 (8-core GPU) |
| Llama 3.1 8B Instruct | Q4_K_M LocalScore | 14.7 llamafile | 15.1 llamafile | even | LocalScore M2 (10-core GPU) · LocalScore M2 (8-core GPU) |
| Llama 3.2 1B Instruct | Q4_K_M LocalScore | 55.6 llamafile | 37.8 llamafile | M2 (10-core GPU) +47% | LocalScore M2 (10-core GPU) · LocalScore M2 (8-core GPU) |
| Qwen2.5 14B Instruct | Q4_K_M LocalScore | 7.5 llamafile | 7.0 llamafile | M2 (10-core GPU) +8% | LocalScore M2 (10-core GPU) · LocalScore M2 (8-core GPU) |
Generation speed in tokens/s, best measured backend for each chip (shown under the number). Backends can differ between the two chips; check them before drawing conclusions. Measured results only, nothing estimated.
Specs
| Apple M2 (10-core GPU) | Apple M2 (8-core GPU) | |
|---|---|---|
| Memory | Unified memory | Unified memory |
| Memory bandwidth | 100 GB/s | 100 GB/s |
All results for the Apple M2 (10-core GPU) · All results for the Apple M2 (8-core GPU) · More comparisons