DavidAU/Qwen3.8-27B-Cold-Fusion-GAIN-V1.1
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, k_proj, o_proj, out_proj, q_proj, up_proj, v_proj changed, in 64 of 64 layers; norms, biases and embeddings are untouched.
Tensors changed: Norms 0/325, Attention 112/560, MLP 192/307, Embeddings 0/5, lm_head 0/1, other 0/1
20 derivatives checked
- DavidAU/Qwen3.8-27B-Cold-Fusion-GAIN-V1.1-NM-DAU-NEO-MAX-MTP-GGUF: Quantization
- Deathsquad10/Qwen3.8B-27B-Mix: Full fine-tune
- andrew-paul/Qwen3.8-27B-Cold-Fusion-GAIN-V1.1-KERN-GGUF: Quantization
- zerodigest/Qwen3.8-27B-Cold-Fusion-GAIN-V1.1-YMQ-GGUF: Quantization
- mradermacher/Qwen3.8-27B-Cold-Fusion-GAIN-V1.1-i1-GGUF: Quantization
- developerjeremylive/Qwen3.8-27B-Cold-Fusion-GAIN-V1.1-NM-DAU-NEO-MAX-MTP-GGUF-etheroi: Quantization
- mradermacher/Qwen3.8-27B-Cold-Fusion-GAIN-V1.1-GGUF: Quantization
- nightmedia/Qwen3.8-27B-Cold-Fusion-GAIN-V1.1-Fable-mxfp8-mlx: Quantization
- nightmedia/Qwen3.8-27B-Fable-Fusion-F711-GAIN-mxfp4-mlx: Quantization
- JC1DA/Qwen3.8-27B-Cold-Fusion-GAIN-V1.1-INT4-Autoround: Quantization
- SeatownSin/Qwen3.8-27B-Cold-Fusion-GAIN-V1.1-NVFP4-W4A16: Quantization
- windowsxp811203/Qwen3.8-27B-Cold-Fusion-GAIN-V1.1-NVFP4: Quantization
- Johneeee/Qwen3.8-27B-Cold-Fusion-GAIN-V1.1-oQ6e-fp16-novision: Quantization
- eemin/Qwen3.8-27B-Cold-Fusion-GAIN-V1.1-NVFP4-W4A16: Quantization
- trithemius/Qwen3.8-27B-Cold-Fusion-GAIN-V1.1-PrismQuant: Quantization
- pyros-vault/Qwen3.8-27B-Cold-Fusion-GAIN-V1.1-oQ6e-mtp: Quantization
- pyros-vault/Qwen3.8-27B-Cold-Fusion-GAIN-V1.1-oQ4e-fp16-mtp: Quantization
- pyros-vault/Qwen3.8-27B-Cold-Fusion-GAIN-V1.1-oQ4e-mtp: Quantization
- gorbatjovy/Qwen3.8-27B-Cold-Fusion-GAIN-V1.1-heretic: Abliterated
- nlpguy/Qwen3.8-27B-Fimi-2: LoRA-style