Error loading Qwen3-Next-80B-A3B: Missing tensor 'blk.0.ssm_in.weight'

#6
by amiii66 - opened

I am encountering a loading error when attempting to run the GGUF quantization of Qwen3-Next-80B-A3B-Thinking-Q4_K_M. The model fails to initialize with the following log:

error loading model: missing tensor 'blk.0.ssm_in.weight'
LM Studio 0.3.39

Same here in llama.cpp (commit 423bee462) - using Qwen3-Coder-Next-UD-Q5_K_XL-00001-of-00002.gguf

Sign up or log in to comment