Cloudflare/clef-flash
LoRA-style vs Qwen/Qwen3.5-9B. 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 32 of 32 layers; norms, biases and embeddings are untouched.
Tensors changed: lm_head 0/1, Embeddings 0/4, MLP 96/208, Attention 152/332, Norms 0/215
26 derivatives checked
- ggml-org/Clef-Flash-GGUF: Quantization
- bartowski/Cloudflare_clef-flash-GGUF: Quantization
- Pitaya1219/Clef-Flash-GGUF: Quantization
- mlx-community/clef-flash-4bit: Quantization
- NANI-Nithin/clef-flash-GGUF: Same weights, converted
- taurusduan/Cloudflare_clef-flash-GGUF: Same weights, converted
- prithivMLmods/clef-flash-GGUF: Same weights, converted
- mlx-community/clef-flash-8bit: Quantization
- Jakevin/clef-flash-mixed-vision-GGUF: Different architecture
- Livesport/clef-flash-GGUF: Different architecture
- saidutta69/clef-flash-heretic: Abliterated
- TrevorJS/clef-flash-mlx-8bit: Quantization
- OlyMahmud/clef-flash-Q4_K_M-GGUF: Quantization
- Jundot/clef-flash-oQ4e: Quantization
- TrevorJS/clef-flash-mlx-4bit: Quantization
- aufklarer/Clef-flash-9B-MLX-4bit: Quantization
- Chopinab/clef-flash-Q2_K-GGUF: Quantization
- alpha-x-ai/clef-flash-NVFP4: Quantization
- prithivMLmods/clef-flash-FP8: Quantization
- thomasgauthier/clef-flash-GGUF: Quantization
- kurcontko/clef-flash-NVFP4: Quantization
- meossistant/clef-flash-4bit: Quantization
- kurcontko/clef-flash-FP8-Dynamic: Quantization
- ApolloRaines/Clef-Flash-Concise-m0.5: Abliterated
- ApolloRaines/Clef-Flash-Truthful-m0.5: Abliterated
- ApolloRaines/Clef-Flash-ReducedSycophancy-m0.5: Abliterated