Cloudflare/clef
LoRA-style vs Qwen/Qwen3.8-27B. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, in_proj_a, in_proj_b, in_proj_qkv, in_proj_z, k_proj, o_proj, out_proj, q_proj, up_proj, v_proj changed, in 24 of 64 layers; norms, biases and embeddings are untouched.
Tensors changed: Norms 0/319, Attention 114/556, MLP 72/304, Embeddings 0/4, lm_head 0/1
11 derivatives checked
- bartowski/Cloudflare_clef-GGUF: Quantization
- ggml-org/Clef-GGUF: Quantization
- taurusduan/Cloudflare_clef-GGUF: Same weights, converted
- abenzerps/Clef-GGUF: Quantization
- mlx-community/clef-4bit: Quantization
- mlx-community/clef-8bit: Quantization
- prithivMLmods/clef-GGUF: Same weights, converted
- simonlehmann/clef-NVFP4: Quantization
- Jundot/clef-oQ4e: Quantization
- trevest/Clef-LoRA-for-Qwen3.8-27B-GGUF: Different architecture
- prithivMLmods/clef-FP8: Quantization