foryoung365/Qwen3-ASR-1.7B-Q4_K_M-GGUF

Quantization (fewer layers) vs Qwen/Qwen3-ASR-1.7B. foryoung365/Qwen3-ASR-1.7B-Q4_K_M-GGUF matches part of Qwen/Qwen3-ASR-1.7B (311 of its 708 tensors; the rest of Qwen/Qwen3-ASR-1.7B has no counterpart). On the shared part: foryoung365/Qwen3-ASR-1.7B-Q4_K_M-GGUF is a GGUF quantization (Q4_K_M) of the same weights: 113 tensors match exactly and 198 within quantization error.

Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/113, MLP 0/84, Attention 0/112

On the Hub · Downloads on Model Pulse