TonyMFJ/Qwen3.8-Flash-Next-WinVulkan-GGUF
Partial fine-tune vs agentionai/Qwen3.8-Flash-Next-ROCmFP4-FAST-imatrix-GGUF. TonyMFJ/Qwen3.8-Flash-Next-WinVulkan-GGUF is a GGUF (Q4_K_M) of different weights. Partial change: 75% of the weights changed (attention, lm_head, mlp, norms), in 0 of 0 layers.
Tensors changed: lm_head 1/1, Norms 7/7, other 0/1, MLP 10/10, Attention 9/9