decadence/ gemma-4:12b-it-qat-mlx-4bit

639 1 week ago

12B — QAT — GGUF — MLX — uncensored — sampling | ~7-13 GB

vision tools thinking audio
... /
json
a88feea3693a · 1.6kB
{
"version": 2,
"weight_format": "mxfp4",
"profile": "MXFP4",
"source_model": {
"name": "gemma-4-12B-it-qat-q4_0-unquantized",
"architecture": "gemma4_unified_text"
},
"has_vision": true,
"has_audio": true,
"quantization": {
"method": "mxfp4",
"quantization_backend": "mx.quantize",
"mode": "mxfp4",
"group_size": 32,
"bits": 4,
"norm_convention": "gemma4_scale_shift_zero",
"tied_embedding": "fp16_passthrough",
"multimodal": "fp16_passthrough_early_fusion_embedders",
"mtp_policy": "none",
"passthrough_bit_widths_used": [
16
],
"passthrough_tensor_count": 349,
"preserved_modal_components": [
"vision_embedder",
"audio_embedder"
]
},
"runtime": {
"total_weight_bytes": 7910289792,
"total_weight_gb": 7.37,
"attention": "hybrid_swa_full_5to1",
"sliding_window": 1024,
"attention_k_eq_v_on_full_layers": true,
"full_attention_layers": [
5,
11,
17,
23,
29,
35,
41,
47
]
},
"capabilities": {
"reasoning_parser": "gemma4",
"tool_parser": "gemma4",
"think_in_template": false,
"supports_tools": true,
"supports_thinking": true,
"family": "gemma4",
"modality": "multimodal",
"modalities": {
"text": true,
"vision": true,
"audio": true,
"video": false
},
"has_vision": true,
"has_audio": true,
"has_video": false,
"cache_type": "kv"
},
"has_video": false,
"modalities": {
"text": true,
"vision": true,
"audio": true,
"video": false
}
}