xxang/AStar-Thought-V2-Qwen3.5-9B

Full fine-tune vs Qwen/Qwen3.5-9B. Fine-tune: 94% of the weights changed, across 32 of 32 layers.

Tensors changed: MLP 96/208, Attention 181/332, Norms 71/215, Embeddings 0/3

2 derivatives checked

• mradermacher/AStar-Thought-V2-Qwen3.5-9B-i1-GGUF: Quantization
• mradermacher/AStar-Thought-V2-Qwen3.5-9B-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse