interfaze-ai/lev
Adapter vs Qwen/Qwen3.5-4B. interfaze-ai/lev is a LoRA adapter, not full weights: rank 32, on down_proj, gate_proj, k_proj, o_proj, q_proj, up_proj, v_proj in 32 layers (128 adapted matrices). It modifies Qwen/Qwen3.5-4B at load time.
1 derivatives checked
- ggml-org/lev-GGUF: Adapter
3 other repos hold byte-for-byte identical weight files.