jaredpalmer/kev-9b
Adapter vs Qwen/Qwen3.5-9B-Base. jaredpalmer/kev-9b is a LoRA adapter, not full weights: rank 16, on down_proj, gate_proj, in_proj_a, in_proj_b, in_proj_qkv, in_proj_z, k_proj, o_proj… in 32 layers (248 adapted matrices). It modifies Qwen/Qwen3.5-9B-Base at load time.
1 derivatives checked
- ggml-org/Kev-9B-GGUF: Adapter