mradermacher/Qwen3-ASR-0.6B-Khmer-GGUF

Same weights, converted (fewer layers) vs seanghay/Qwen3-ASR-0.6B-Khmer. mradermacher/Qwen3-ASR-0.6B-Khmer-GGUF matches part of seanghay/Qwen3-ASR-0.6B-Khmer (310 of its 611 tensors; the rest of seanghay/Qwen3-ASR-0.6B-Khmer has no counterpart). On the shared part: Same weights stored in another dtype (BF16 → F16/F32).

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

On the Hub · Downloads on Model Pulse