YOYO-AI/Qwen3-30B-A3B-CoderThinking-YOYO-linear

Full fine-tune vs Qwen/Qwen3-30B-A3B-Thinking-2507. Fine-tune: 100% of the weights changed, across 48 of 48 layers.

Tensors changed: Norms 193/193, MLP 2613/2613, Attention 192/192

2 derivatives checked

• mradermacher/Qwen3-30B-A3B-CoderThinking-YOYO-linear-i1-GGUF: Quantization
• mradermacher/Qwen3-30B-A3B-CoderThinking-YOYO-linear-GGUF: Quantization

On the Hub · Downloads on Model Pulse