Llama 4 Maverick 17B-128E Instruct: What PC You Need

Llama 4 Maverick 17B-128E Instruct needs about 280.9GB at Q4. No PC we track with a verified price runs it comfortably.

What Llama 4 Maverick 17B-128E Instruct asks of a PC

Llama 4 Maverick 17B-128E Instruct is a mixture-of-experts model with 400B parameters, 17B of them active per token. Every expert has to stay resident, so its memory need follows the 400B figure, not the 17B one.

At 8K context Llama 4 Maverick 17B-128E Instruct takes 280.9GB at Q4, 381.5GB at Q6 and 490.7GB at Q8. Stepping from Q4 to Q8 costs 209.8GB.

Llama 4 Maverick 17B-128E Instruct accepts up to 1,000,000 tokens of context, and a long context adds memory: at 128K and Q4 it takes 310.9GB.

Llama 4 Maverick 17B-128E Instruct was released in 2025-04.

0 tracked PCs with a verified price run Llama 4 Maverick 17B-128E Instruct comfortably and 0 run it with limits.

Memory requirements for Llama 4 Maverick 17B-128E Instruct

Quantization	8K context	32K context	128K context
Q4	280.9GB	286.9GB	310.9GB
Q5	329.8GB	335.8GB	359.8GB
Q6	381.5GB	387.5GB	411.5GB
Q8	490.7GB	496.7GB	520.8GB

Parameters: 400B
Active per token: 17B
Context window: 1,000,000
Source: Model card

Cheapest PCs that run Llama 4 Maverick 17B-128E Instruct comfortably

None we track with a verified price.

Cheapest PCs that run Llama 4 Maverick 17B-128E Instruct with limitations