MODEL / ALIBABA QWEN

Qwen3 30B-A3B

All 30B parameters must be stored even though roughly 3B are active per token.

総パラメータ30B3B active / token

推定メモリ~22.5 GBGGUF Q4 MoE

コンテキスト32K+Text

LicenseApache-2.0100+ languages and dialects

01 / 用途

02 / 推定適合

03 / 実行環境

OllamaEasyollama run qwen3:30b-a3b

llama.cppMediumllama-server -m /path/to/verified-model.gguf --port 8080

LM StudioEasySearch “Qwen/Qwen3-30B-A3B” → Load a Q4/INT4 build → Start Server

MLX LMMediummlx_lm.chat --model Qwen/Qwen3-30B-A3B

注意

All 30B parameters must be stored even though roughly 3B are active per token.