RunsOnMy ModelsGPUs & MacsBy VRAMAPI pricesVRAM calculatorNew

Can SmolVLM2 500M Video Instruct run on the Mac M4 32GB?

Yes. At 8K context the best fit is F16, using about 1.6 GB of the 21.4 GB usable on the Mac M4 32GB. Estimated speed ceiling: ~146 tokens/sec.

Memory by quant and context

Quant4K ctx8K ctxMax tok/s (est., 8K)
2-bit (Q2_K)0.8 GB Fits1.0 GB Fits631
3-bit (Q3_K_M)0.9 GB Fits1.0 GB Fits484
4-bit (Q4_K_M)0.9 GB Fits1.1 GB Fits387
5-bit (Q5_K_M)1.0 GB Fits1.1 GB Fits332
6-bit (Q6_K)1.0 GB Fits1.2 GB Fits288
8-bit (Q8_0) •1.1 GB Fits1.2 GB Fits275
16-bit (F16/BF16) •1.4 GB Fits1.6 GB Fits146

• real GGUF file size from ggml-org/SmolVLM2-500M-Video-Instruct-GGUF. Speed is a bandwidth ceiling (120 GB/s), not a benchmark. Methodology.

More

SmolVLM2 500M Video Instruct on every GPU and Mac · Every model that runs on the Mac M4 32GB · Apple Silicon guide · Try other settings in the calculator

Data: Hugging Face Hub (HuggingFaceTB/SmolVLM2-500M-Video-Instruct), Apple specs. Data as of 2026-11-17.