Image-Text-to-Text
MLX
Safetensors
deepseekocr_2
mlx-vlm
axquant
axq
deepseek
ocr
vision
development
conversational
6-bit
Instructions to use AutomatosX/AX-DeepSeek-OCR-2-MLX-AXQ-6bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use AutomatosX/AX-DeepSeek-OCR-2-MLX-AXQ-6bit with MLX:
# Make sure mlx-vlm is installed # pip install --upgrade mlx-vlm from mlx_vlm import load, generate from mlx_vlm.prompt_utils import apply_chat_template from mlx_vlm.utils import load_config # Load the model model, processor = load("AutomatosX/AX-DeepSeek-OCR-2-MLX-AXQ-6bit") config = load_config("AutomatosX/AX-DeepSeek-OCR-2-MLX-AXQ-6bit") # Prepare input image = ["http://images.cocodataset.org/val2017/000000039769.jpg"] prompt = "Describe this image." # Apply chat template formatted_prompt = apply_chat_template( processor, config, prompt, num_images=1 ) # Generate output output = generate(model, processor, formatted_prompt, image) print(output) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
| { | |
| "created_at": "2026-08-13T23:54:54.123547Z", | |
| "plan_sha256": "552b7102998fda89c850dcb5acccef68131f1f8235249c171d7a2e70cdc90e06", | |
| "records": [ | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.embed_tokens", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 32, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.0.mlp.down_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 32, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.0.mlp.gate_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 32, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.0.mlp.up_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.0.self_attn.k_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.0.self_attn.o_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.0.self_attn.q_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.0.self_attn.v_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.1.mlp.shared_experts.down_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.1.mlp.shared_experts.gate_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.1.mlp.shared_experts.up_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.1.mlp.switch_mlp.down_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.1.mlp.switch_mlp.gate_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.1.mlp.switch_mlp.up_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.1.self_attn.k_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.1.self_attn.o_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.1.self_attn.q_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.1.self_attn.v_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.2.mlp.shared_experts.down_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.2.mlp.shared_experts.gate_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.2.mlp.shared_experts.up_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.2.mlp.switch_mlp.down_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.2.mlp.switch_mlp.gate_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.2.mlp.switch_mlp.up_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.2.self_attn.k_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.2.self_attn.o_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.2.self_attn.q_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.2.self_attn.v_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.3.mlp.shared_experts.down_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.3.mlp.shared_experts.gate_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.3.mlp.shared_experts.up_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.3.mlp.switch_mlp.down_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.3.mlp.switch_mlp.gate_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.3.mlp.switch_mlp.up_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.3.self_attn.k_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.3.self_attn.o_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.3.self_attn.q_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.3.self_attn.v_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.4.mlp.shared_experts.down_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.4.mlp.shared_experts.gate_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.4.mlp.shared_experts.up_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.4.mlp.switch_mlp.down_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.4.mlp.switch_mlp.gate_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.4.mlp.switch_mlp.up_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.4.self_attn.k_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.4.self_attn.o_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.4.self_attn.q_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.4.self_attn.v_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.5.mlp.shared_experts.down_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.5.mlp.shared_experts.gate_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.5.mlp.shared_experts.up_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.5.mlp.switch_mlp.down_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.5.mlp.switch_mlp.gate_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.5.mlp.switch_mlp.up_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.5.self_attn.k_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.5.self_attn.o_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.5.self_attn.q_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.5.self_attn.v_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.6.mlp.shared_experts.down_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.6.mlp.shared_experts.gate_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.6.mlp.shared_experts.up_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.6.mlp.switch_mlp.down_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.6.mlp.switch_mlp.gate_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.6.mlp.switch_mlp.up_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.6.self_attn.k_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.6.self_attn.o_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.6.self_attn.q_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.6.self_attn.v_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.7.mlp.shared_experts.down_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.7.mlp.shared_experts.gate_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.7.mlp.shared_experts.up_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.7.mlp.switch_mlp.down_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.7.mlp.switch_mlp.gate_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.7.mlp.switch_mlp.up_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.7.self_attn.k_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.7.self_attn.o_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.7.self_attn.q_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.7.self_attn.v_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.8.mlp.shared_experts.down_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.8.mlp.shared_experts.gate_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.8.mlp.shared_experts.up_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.8.mlp.switch_mlp.down_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.8.mlp.switch_mlp.gate_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.8.mlp.switch_mlp.up_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.8.self_attn.k_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.8.self_attn.o_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.8.self_attn.q_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.8.self_attn.v_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.9.mlp.switch_mlp.gate_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.9.mlp.switch_mlp.up_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.9.self_attn.k_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.9.self_attn.o_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.9.self_attn.q_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.9.self_attn.v_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 32, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.10.mlp.shared_experts.down_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 32, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.10.mlp.shared_experts.gate_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 32, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.10.mlp.shared_experts.up_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.10.mlp.switch_mlp.down_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.10.mlp.switch_mlp.gate_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.10.mlp.switch_mlp.up_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.10.self_attn.k_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.10.self_attn.o_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.10.self_attn.q_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.10.self_attn.v_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 32, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.11.mlp.shared_experts.down_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 32, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.11.mlp.shared_experts.gate_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 32, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.11.mlp.shared_experts.up_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.11.mlp.switch_mlp.down_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.11.mlp.switch_mlp.gate_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.11.mlp.switch_mlp.up_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.11.self_attn.k_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.11.self_attn.o_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.11.self_attn.q_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 8, | |
| "fallback": false, | |
| "group_size": 64, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.11.self_attn.v_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 32, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.9.mlp.shared_experts.down_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 32, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.9.mlp.shared_experts.gate_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 32, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.9.mlp.shared_experts.up_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| }, | |
| { | |
| "bits": 6, | |
| "fallback": false, | |
| "group_size": 32, | |
| "metadata": {}, | |
| "method": "affine", | |
| "module_path": "language_model.model.layers.9.mlp.switch_mlp.down_proj", | |
| "note": "mlx-vlm affine packing", | |
| "success": true | |
| } | |
| ], | |
| "schema_version": "axquant.quantizer-execution.v1" | |
| } | |