mradermacher/NOESIS-Qwen3-TTS-1.7B-Talker-Combo-LLM2-ASR2-BF16-GGUF
Same weights, converted (fewer layers) vs AMAImedia/NOESIS-Qwen3-TTS-1.7B-Talker-Combo-LLM2-ASR2-BF16. mradermacher/NOESIS-Qwen3-TTS-1.7B-Talker-Combo-LLM2-ASR2-BF16-GGUF matches part of AMAImedia/NOESIS-Qwen3-TTS-1.7B-Talker-Combo-LLM2-ASR2-BF16 (309 of its 480 tensors; the rest of AMAImedia/NOESIS-Qwen3-TTS-1.7B-Talker-Combo-LLM2-ASR2-BF16 has no counterpart). On the shared part: Same weights stored in another dtype (BF16 → F16/F32), plus 2 new tensors.
Tensors changed: Norms 0/113, MLP 0/84, Attention 0/112