SPR Qwen3.5-9B DoRa VRAM-safe GGUF Pack
Inference-ready GGUF quantizations exported from the SPR adapter in this repo and
merged back to full model weights.
Included files
spr_qwen3_5_9b_dora_vramsafe_bf16.ggufspr_qwen3_5_9b_dora_vramsafe_q4_k_m.ggufspr_qwen3_5_9b_dora_vramsafe_q6_k.ggufspr_qwen3_5_9b_dora_vramsafe_q8_0.gguf
Notes
Use with llama.cpp compatible runtimes (LM Studio / llama.cpp CLI). If you use a
single-stop token setup, keep ###STOP### active.
