Upload twin-tower VLM using push_to_hub

README.md

@@ -0,0 +1,54 @@
+
+---
+library_name: nanovlm
+license: mit
+pipeline_tag: image-text-to-text
+tags:
+  - vision-language
+  - multimodal
+  - research
+  - twin-tower
+---
+
+**Twin-Tower VLM** is a vision-language model based on the twin-tower architecture. This model uses a separate vision tower to process images and generate per-layer contexts, which are then integrated with a frozen language tower for text generation.
+
+## Architecture
+
+The twin-tower architecture consists of:
+
+1. **Vision Tower**: Processes images through vision encoder → modality projector → decoder layers to create per-layer contexts
+2. **Language Tower**: Frozen language model that receives vision contexts and generates text
+
+## Key Features
+
+- **Twin-Tower Design**: Separate processing of vision and language with per-layer context integration
+- **Frozen Language Tower**: Language model parameters are frozen, gradients flow through vision contexts
+- **Per-Layer Contexts**: Vision tower generates contexts for each language model layer
+- **Efficient Training**: Only vision tower components are trainable
+
+## Usage
+
+```python
+from twin_tower import VisionLanguageTwinTowerModel
+from config import VLMConfig
+
+# Load the model
+cfg = VLMConfig()
+model = VisionLanguageTwinTowerModel.from_pretrained(cfg)
+
+# Generate text from image
+from PIL import Image
+image = Image.open("your_image.jpg")
+result = model.generate_from_text("What is in this image?", image)
+print(result)
+```
+
+## Model Details
+
+- **Base Model**: patrickamadeus/nanoVLM-230M-8k-twin-maxxing-3000
+- **Architecture**: Twin-Tower VLM
+- **Vision Encoder**: SigLIP-based
+- **Language Model**: SmolLM2-based
+- **Parameters**: ~230M total (vision tower trainable, language tower frozen)
+
+For more information, check out the base nanoVLM model: https://huggingface.co/lusxvr/nanoVLM-222M.

config.json

@@ -0,0 +1,106 @@
+{
+    "vit_hidden_dim": 768,
+    "vit_inter_dim": 3072,
+    "vit_patch_size": 16,
+    "vit_img_size": 512,
+    "vit_n_heads": 12,
+    "vit_dropout": 0.0,
+    "vit_n_blocks": 12,
+    "vit_ln_eps": 1e-06,
+    "vit_cls_flag": false,
+    "vit_model_type": "google/siglip2-base-patch16-512",
+    "lm_hidden_dim": 576,
+    "lm_inter_dim": 1536,
+    "lm_rms_eps": 1e-05,
+    "lm_re_base": 100000,
+    "lm_max_position_embeddings": 8192,
+    "lm_base_vocab_size": 49152,
+    "extra_token_amount": 66,
+    "lm_vocab_size": 49218,
+    "lm_n_heads": 9,
+    "lm_n_kv_heads": 3,
+    "lm_dropout": 0.0,
+    "lm_n_blocks": 30,
+    "lm_attn_scaling": 1.0,
+    "lm_max_length": 256,
+    "lm_use_tokens": false,
+    "lm_tie_weights": true,
+    "lm_model_type": "HuggingFaceTB/SmolLM2-135M-Instruct",
+    "lm_tokenizer": "HuggingFaceTB/SmolLM2-360M-Instruct",
+    "lm_chat_template": "{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
+    "mp_pixel_shuffle_factor": 4,
+    "mp_image_token_length": 64,
+    "max_img_size": 512,
+    "resize_to_max_side_len": false,
+    "vlm_extra_tokens": {
+        "image_token": "<|image|>",
+        "global_image_token": "<|global_image|>",
+        "r1c1": "<row_1_col_1>",
+        "r1c2": "<row_1_col_2>",
+        "r1c3": "<row_1_col_3>",
+        "r1c4": "<row_1_col_4>",
+        "r1c5": "<row_1_col_5>",
+        "r1c6": "<row_1_col_6>",
+        "r1c7": "<row_1_col_7>",
+        "r1c8": "<row_1_col_8>",
+        "r2c1": "<row_2_col_1>",
+        "r2c2": "<row_2_col_2>",
+        "r2c3": "<row_2_col_3>",
+        "r2c4": "<row_2_col_4>",
+        "r2c5": "<row_2_col_5>",
+        "r2c6": "<row_2_col_6>",
+        "r2c7": "<row_2_col_7>",
+        "r2c8": "<row_2_col_8>",
+        "r3c1": "<row_3_col_1>",
+        "r3c2": "<row_3_col_2>",
+        "r3c3": "<row_3_col_3>",
+        "r3c4": "<row_3_col_4>",
+        "r3c5": "<row_3_col_5>",
+        "r3c6": "<row_3_col_6>",
+        "r3c7": "<row_3_col_7>",
+        "r3c8": "<row_3_col_8>",
+        "r4c1": "<row_4_col_1>",
+        "r4c2": "<row_4_col_2>",
+        "r4c3": "<row_4_col_3>",
+        "r4c4": "<row_4_col_4>",
+        "r4c5": "<row_4_col_5>",
+        "r4c6": "<row_4_col_6>",
+        "r4c7": "<row_4_col_7>",
+        "r4c8": "<row_4_col_8>",
+        "r5c1": "<row_5_col_1>",
+        "r5c2": "<row_5_col_2>",
+        "r5c3": "<row_5_col_3>",
+        "r5c4": "<row_5_col_4>",
+        "r5c5": "<row_5_col_5>",
+        "r5c6": "<row_5_col_6>",
+        "r5c7": "<row_5_col_7>",
+        "r5c8": "<row_5_col_8>",
+        "r6c1": "<row_6_col_1>",
+        "r6c2": "<row_6_col_2>",
+        "r6c3": "<row_6_col_3>",
+        "r6c4": "<row_6_col_4>",
+        "r6c5": "<row_6_col_5>",
+        "r6c6": "<row_6_col_6>",
+        "r6c7": "<row_6_col_7>",
+        "r6c8": "<row_6_col_8>",
+        "r7c1": "<row_7_col_1>",
+        "r7c2": "<row_7_col_2>",
+        "r7c3": "<row_7_col_3>",
+        "r7c4": "<row_7_col_4>",
+        "r7c5": "<row_7_col_5>",
+        "r7c6": "<row_7_col_6>",
+        "r7c7": "<row_7_col_7>",
+        "r7c8": "<row_7_col_8>",
+        "r8c1": "<row_8_col_1>",
+        "r8c2": "<row_8_col_2>",
+        "r8c3": "<row_8_col_3>",
+        "r8c4": "<row_8_col_4>",
+        "r8c5": "<row_8_col_5>",
+        "r8c6": "<row_8_col_6>",
+        "r8c7": "<row_8_col_7>",
+        "r8c8": "<row_8_col_8>"
+    },
+    "vlm_load_backbone_weights": true,
+    "vlm_checkpoint_path": "lusxvr/nanoVLM-230M-8k",
+    "hf_repo_name": "nanoVLM"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9ffe689bdf230a1b510499dc847ae155bf0ac0392a9f1916828ee94b1a4de2f
+size 1450549888