KaraKaraWitch/BiggerCoQ-Qwen3-10b

Partial fine-tune vs KaraKaraWitch/CavesOfQwen3-8b. Partial change: 66% of the weights changed (attention, mlp, norms), in 28 of 36 layers.

Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 112/145, MLP 84/108, Attention 112/144

2 derivatives checked

• mradermacher/BiggerCoQ-Qwen3-10b-i1-GGUF: Quantization
• mradermacher/BiggerCoQ-Qwen3-10b-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse