On a 16 GB Mac, 12 catalog models run fully on the GPU at an 8,192-token context. The most capable is Mistral-Small-3.2-24B-Instruct-2506 at Q4_K_M (needs ~15.2 GiB). Pick a smaller model or a lower quant for more headroom.
Figures assume a 16 GB unified-memory Mac (one pool shared by CPU and GPU), judged at an 8,192-token context. Speeds depend on the specific chip — see the GPU pages.
Mistral-Small-3.2-24B-Instruct-2506 at Q4_K_M · 24B params · needs ~15.2 GiB
One command to run it (llama.cpp):
llama-server -m mistralai_Mistral-Small-3.2-24B-Instruct-2506-Q4_K_M.gguf -c 8192 -ngl 999| Model | Sweet-spot quant | Fits in |
|---|---|---|
| Mistral-Small-3.2-24B-Instruct-2506 mistralai | Q4_K_M Runs fully on GPU | ~15.2 GiB |
| Devstral-Small-2-24B-Instruct-2512 mistralai | IQ4_XS Runs fully on GPU | ~14.9 GiB |
| gpt-oss-20b openai | F16 Runs fully on GPU | ~15 GiB |
| Kimi-VL-A3B-Instruct moonshotai | Q4_K_M Runs fully on GPU | ~13 GiB |
| DeepSeek-R1-Distill-Qwen-14B deepseek-ai | Q4_K_M Runs fully on GPU | ~11.2 GiB |
| Qwen3-14B Qwen | Q4_K_M Runs fully on GPU | ~11 GiB |
| Phi-4-reasoning microsoft | Q4_K_M Runs fully on GPU | ~11.4 GiB |
| phi-4 microsoft | Q4_K_M Runs fully on GPU | ~11.2 GiB |
| gemma-4-12B-it google | Q8_0 Runs fully on GPU | ~13.5 GiB |
| Qwen3-8B Qwen | Q8_0 Runs fully on GPU | ~10.6 GiB |
| DeepSeek-R1-Distill-Llama-8B deepseek-ai | Q8_0 Runs fully on GPU | ~10.3 GiB |
| gemma-4-E4B-it google | Q8_0 Runs fully on GPU | ~8.9 GiB |
Derived live from the fit engine + catalog at an 8,192-token context. "Fits in" is the modelled VRAM the sweet-spot quant needs (weights + KV cache + overhead). Speed depends on your specific card — check a GPU page or the calculator.
selected to compare · pick at least 2