← Back to Models
Alibabamoe
Qwen3 30B-A3B
30.5B total / 3.3B active · load estimate uses total • 18.6GB VRAM (Q4 estimate) • 131,072 context
Estimate · Q4_K_M · 4k context · batch 1. For dense 7B–70B this sits ~5–8% above the Q4_K_M GGUF file. Not peak runtime VRAM and not a measured bench.
Specifications
Parameters
30.5B total / 3.3B active · load estimate uses total
VRAM (Q4 est.)
18.6 GB
VRAM (FP16)
61 GB
Context Window
131,072
Architecture
moe
License
Apache-2.0
Finding cloud alternatives...
Run Qwen3 30B-A3B on RTX 4000 Ada
~18.6GB VRAM needed at Q4. RTX 4000 Ada has 20GB — buy hardware or rent cloud.
We may earn a commission from cloud and hardware partners at no extra cost to you.
Find the cheapest GPU that can run Qwen3 30B-A3B
Compatibility Lab — check every GPU × quantization combination