Rent a GPU

Qwen3 235B-A22B hardware requirements

235.1B parameters (22B active, mixture of experts). Needs about 144.7 GB at Q4_K_M with 8K context.

Parameters235B22B active (MoE)
Q4_K_M · 8K144.7 GBthe usual download
FP16 · 8K472.4 GBfull precision
Runs well on1/51GPUs 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 quality472.4 GB477.1 GB
Q8_0Near-lossless252 GB256.7 GB
Q6_KExcellent195 GB199.7 GB
Q5_K_MVery good169.4 GB174.1 GB
Q4_K_MGood, most popular144.7 GB149.4 GB
Q3_K_MNoticeably degraded117.1 GB121.8 GB
Q2_KHeavily degraded90.3 GB95.1 GB

Run it with Ollama

Downloads Q4_K_M by default.

ollama run qwen3:235b

Model card

Qwen/Qwen3-235B-A22B on Hugging Face: licence, prompt format and GGUF builds.

Run it in the cloud

Run Qwen3 235B-A22B on a rented GPU today.

It needs about 144.7 GB at Q4_K_M: a rented 2× A100 / H100 runs it for about $2.22 to $7.98 an hour.

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 235B-A22B fills at the best quant that fits.

1 Runs well

NameMemory usedBest quantSpeed

3 Fits with heavy quantization

NameMemory usedBest quantSpeed

47 Too big

NameMemory usedBest quantSpeed

Other Qwen models