AMD Instinct MI250X~72 tok/s (Q4)

deepseek-ai/DeepSeek-Coder-V2-Instruct-0724 speed on AMD Instinct MI250X

Quantization-specific throughput and VRAM requirements for deepseek-ai/DeepSeek-Coder-V2-Instruct-0724 running on AMD Instinct MI250X.

Speed Snapshot

Topline estimate from compatibility data

Modeldeepseek-ai/DeepSeek-Coder-V2-Instruct-0724

GPUAMD Instinct MI250X

Q4 speed72 tok/s

Q4 VRAM required115GB

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	115GB	128GB	72 tok/s	✅ Fits
Q8	231GB	128GB	54 tok/s	❌ Not recommended
FP16	461GB	128GB	29 tok/s	❌ Not recommended