Apple M2 Ultra~39 tok/s (Q4)

deepseek-ai/deepseek-coder-33b-instruct speed on Apple M2 Ultra

Quantization-specific throughput and VRAM requirements for deepseek-ai/deepseek-coder-33b-instruct running on Apple M2 Ultra.

Speed Snapshot

Topline estimate from compatibility data

Modeldeepseek-ai/deepseek-coder-33b-instruct

GPUApple M2 Ultra

Q4 speed39 tok/s

Q4 VRAM required17GB

Data Source

Calculation and benchmark status

Speed values come from the compatibility dataset (`estimatedTokensPerSec`) and are sorted by quantization.

For full verdict logic and alternate GPUs, see the canonical compatibility page.

Quantization Speed Table

Quantization	VRAM needed	VRAM available	Speed	Verdict
Q4	17GB	192GB	39 tok/s	✅ Fits
Q8	34GB	192GB	26 tok/s	✅ Fits
FP16	68GB	192GB	14 tok/s	✅ Fits