YOYO-AI/Qwen3-30B-A3B-Deepseek-Distill-Instruct-2507

Partial fine-tune vs Qwen/Qwen3-30B-A3B-Instruct-2507. Partial change: 85% of the weights changed (attention, mlp, norms), in 48 of 48 layers.

Tensors changed: Norms 165/193, MLP 2280/2613, Attention 143/192

2 derivatives checked

• mradermacher/Qwen3-30B-A3B-Deepseek-Distill-Instruct-2507-i1-GGUF: Quantization
• mradermacher/Qwen3-30B-A3B-Deepseek-Distill-Instruct-2507-GGUF: Quantization

On the Hub · Downloads on Model Pulse