chrullis/Qwen3-ASR-1.7B-Somali

Partial fine-tune vs Qwen/Qwen3-ASR-1.7B. Partial change: 85% of the weights changed (attention, mlp, other), in 28 of 28 layers.

Tensors changed: other 3/11, MLP 132/180, Norms 0/211, Attention 208/304, Embeddings 0/1

1 derivatives checked

• mradermacher/Qwen3-ASR-1.7B-Somali-GGUF: Same weights, converted (fewer layers)

On the Hub · Downloads on Model Pulse