LLMLocal inference
Qwen3 72B
VRAM requirements to run Qwen3 72B locally at each quantization level. Find the cheapest GPU that fits below.
Q4 VRAM
42 GB
Q8 VRAM
78 GB
FP16 VRAM
144 GB
Context window
128 k tokens
01 // GPUs that can run Qwen3 72B
Cheapest compatible hardware by quantization
Sorted cheapest first. All prices are approximate street prices.
Q4Q4_K_M (4-bit)
needs ≥42 GB VRAMGPUVRAMPriceTier
Q8Q8_0 (8-bit)
needs ≥78 GB VRAMGPUVRAMPriceTier
FP16FP16 (full precision)
needs ≥144 GB VRAMGPUVRAMPriceTier
02 // Frequently asked
Qwen3 72B GPU questions
How much VRAM does Qwen3 72B need?
Qwen3 72B requires approximately 42GB VRAM at Q4 quantization, 78GB at Q8, or 144GB at full FP16 precision. Q4 is the most practical choice for consumer hardware.
What is the cheapest GPU to run Qwen3 72B?
The cheapest single GPU that fits Qwen3 72B at Q4 is the Apple M4 Pro (48GB VRAM, ~$2,499). At Q4 you need at least 42GB.
Can I run Qwen3 72B at FP16?
Qwen3 72B at FP16 requires 144GB VRAM — well beyond any single consumer GPU. FP16 is only practical on multi-GPU server configurations. Q4 (42GB) or Q8 (78GB) are the realistic options.
What quantization is best for Qwen3 72B?
Q4_K_M (42GB) offers the best hardware compatibility and still produces high-quality output. Q8_0 (78GB) is better for tasks needing higher accuracy at the cost of needing more VRAM. FP16 (144GB) is only practical on very high-end workstation hardware.