mradermacher/tamil-asr-qwen3-GGUF

Same weights, converted (fewer layers) vs lemuralabs/tamil-asr-qwen3. mradermacher/tamil-asr-qwen3-GGUF matches part of lemuralabs/tamil-asr-qwen3 (310 of its 707 tensors; the rest of lemuralabs/tamil-asr-qwen3 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