Rent a GPU

Qwen3.8 2.4T-A95B hardware requirements

2446B parameters (95B active, mixture of experts). Needs about 1484.3 GB at Q4_K_M with 8K context.

Parameters2446B95B active (MoE)
Q4_K_M · 8K1484.3 GBthe usual download
FP16 · 8K4893.4 GBfull precision
Runs well on0/51GPUs and Macs we track

Memory needed at each quantization

Weights, KV cache and runtime overhead, at 8K and 32K tokens of context.

QuantQuality8K context32K context
FP16Full quality4893.4 GB4895.7 GB
Q8_0Near-lossless2600.2 GB2602.6 GB
Q6_KExcellent2007.1 GB2009.4 GB
Q5_K_MVery good1741.1 GB1743.4 GB
Q4_K_MGood, most popular1484.3 GB1486.6 GB
Q3_K_MNoticeably degraded1196.9 GB1199.2 GB
Q2_KHeavily degraded918.6 GB920.9 GB

Model card

Qwen/Qwen3.8-2.4T-A95B on Hugging Face: licence, prompt format and GGUF builds.

Run it in the cloud

Run Qwen3.8 2.4T-A95B on a rented GPU today.

It needs about 1484.3 GB at Q4_K_M: a rented 8× B300 runs it for about $71.92 an hour. Compare with Claude and APIs or see how it scores against them.

RunPodRent a GPU on RunPodVast.aiRent a GPU on Vast.ai

Every GPU and Mac checked

Sorted by memory. Bars show how much of each card Qwen3.8 2.4T-A95B fills at the best quant that fits.

51 Too big

NameMemory usedBest quantSpeed

Other Qwen models