Locutusque/Esmeralda-Qwen3.6-35B-A3B

Partial fine-tune vs Qwen/Qwen3.6-35B-A3B. Partial change: 10% of the weights changed (attention, mlp), in 40 of 40 layers.

Tensors changed: Embeddings 0/5, Attention 190/392, MLP 168/399, Norms 0/247, lm_head 0/1, other 0/1

2 derivatives checked

• mradermacher/Esmeralda-Qwen3.6-35B-A3B-i1-GGUF: Quantization
• mradermacher/Esmeralda-Qwen3.6-35B-A3B-GGUF: Quantization

On the Hub · Downloads on Model Pulse