DavidAU/Qwen3-Zero-Coder-Reasoning-V2-0.8B
Partial fine-tune (fewer layers) vs Qwen/Qwen3-0.6B. Qwen/Qwen3-0.6B matches part of DavidAU/Qwen3-Zero-Coder-Reasoning-V2-0.8B (310 of its 464 tensors; the rest of DavidAU/Qwen3-Zero-Coder-Reasoning-V2-0.8B has no counterpart). On the shared part: Partial change: 45% of the weights changed (attention, mlp, norms), in 17 of 28 layers.
Tensors changed: Embeddings 0/1, Norms 68/113, MLP 51/84, Attention 68/112
3 derivatives checked
- DavidAU/Qwen3-Zero-Coder-Reasoning-V2-0.8B-NEO-EX-GGUF: Same weights, converted
- mradermacher/Qwen3-Zero-Coder-Reasoning-V2-0.8B-i1-GGUF: Quantization
- mradermacher/Qwen3-Zero-Coder-Reasoning-V2-0.8B-GGUF: Same weights, converted