RunsOnMy ModelsGPUs & MacsBy VRAMAPI pricesVRAM calculatorNew

Can PaddleOCR VL 1.6 run on the Mac M3 Pro 36GB?

Yes. At 8K context the best fit is F16, using about 2.5 GB of the 24.1 GB usable on the Mac M3 Pro 36GB. Estimated speed ceiling: ~78 tokens/sec.

Newer compressed-attention design: long-context figures are an upper bound.

Memory by quant and context

Quant4K ctx8K ctx32K ctx128K ctxMax tok/s (est., 8K)
2-bit (Q2_K)1.0 GB Fits1.1 GB Fits1.5 GB Fits3.2 GB Fits417
3-bit (Q3_K_M)1.1 GB Fits1.2 GB Fits1.6 GB Fits3.3 GB Fits320
4-bit (Q4_K_M)1.2 GB Fits1.3 GB Fits1.7 GB Fits3.4 GB Fits256
5-bit (Q5_K_M)1.3 GB Fits1.4 GB Fits1.8 GB Fits3.5 GB Fits220
6-bit (Q6_K)1.4 GB Fits1.5 GB Fits1.9 GB Fits3.6 GB Fits191
8-bit (Q8_0)1.6 GB Fits1.7 GB Fits2.1 GB Fits3.8 GB Fits147
16-bit (F16/BF16)2.5 GB Fits2.5 GB Fits2.9 GB Fits4.6 GB Fits78

• real GGUF file size. Speed is a bandwidth ceiling (150 GB/s), not a benchmark. Methodology.

More

PaddleOCR VL 1.6 on every GPU and Mac · Every model that runs on the Mac M3 Pro 36GB · Apple Silicon guide · Try other settings in the calculator

Data: Hugging Face Hub (PaddlePaddle/PaddleOCR-VL-1.6), Apple specs. Data as of 2026-11-17.