Qwen3 30B-A3B Instruct

Running Qwen3 30B-A3B Instruct locally

At Q4_K_M and 4K of context, Qwen3 30B-A3B Instruct needs about 19.7 GB — 17.6 GB of weights, 0.38 GB of KV cache and a 1.8 GB activation buffer. The smallest cards in this index that clear that comfortably are 24 GB ones — 4 of them, from the RTX 3090 up — and 31 of the 78 cards here do. These are calculated figures, not measurements: the estimate stops counting a card as comfortable at 88% of its VRAM, which is roughly the room a desktop session needs.

What longer context costs

Going from 4K to 32K adds about 2.62 GB, taking the total to 22.6 GB. Weights do not move with context — only the KV cache does, and it grows linearly, so this is the number to watch when planning for long documents. The model's native window is 40K; holding all of it at Q4_K_M would need about 23 GB.

Which build to download

This index tracks 2 formats for it — GGUF, AWQ — across 3 levels. Q5_K_M carries the lowest published perplexity loss at 1.1%. The fastest level measured here is AWQ INT4 at 118 tok/s on an RTX 4090, batch 1. GGUF runs on llama.cpp and Ollama across NVIDIA, AMD and Apple silicon; AWQ and GPTQ target vLLM on CUDA and ROCm; EXL2 is ExLlamaV2 — CUDA only, and now archived, so local use rather than serving.

Common questions

How much VRAM does Qwen3 30B-A3B Instruct need?

About 19.7 GB at Q4_K_M with 4K of context and batch 1, rising to roughly 22.6 GB at 32K. That figure is weights plus KV cache plus a 10% activation buffer, calculated from the model's architecture rather than measured on a card.

Will Qwen3 30B-A3B Instruct run on a 24GB GPU?

Yes — at Q4_K_M and 4K context it needs about 19.7 GB, which leaves 4.3 GB spare on a RTX 3090, or on any of the other 24 GB cards here (4 in all). That is the smallest size in this index that clears it comfortably; 31 of 78 do.

Which quantization of Qwen3 30B-A3B Instruct should I use?

Q5_K_M has the lowest published quality loss (1.1%), and Q4_K_M is the level most people run. All 3 levels in the index are Q4_K_M, Q5_K_M, AWQ INT4. Note this is a mixture-of-experts model — all parameters must be resident even though only a fraction are active per token, so the memory cost follows the total, not the active count.