Update tiny models for SmolVLMForConditionalGeneration

chat_template.jinja

@@ -0,0 +1,2 @@
+<|im_start|>{% for message in messages %}{{message['role'] | capitalize}}{% if message['content'][0]['type'] == 'image' %}{{':'}}{% else %}{{': '}}{% endif %}{% for line in message['content'] %}{% if line['type'] == 'text' %}{{line['text']}}{% elif line['type'] == 'image' %}{{ '<image>' }}{% endif %}{% endfor %}<end_of_utterance>
+{% endfor %}{% if add_generation_prompt %}{{ 'Assistant:' }}{% endif %}

config.json

@@ -0,0 +1,58 @@
+{
+  "architectures": [
+    "SmolVLMForConditionalGeneration"
+  ],
+  "dtype": "float32",
+  "image_token_id": 49190,
+  "model_type": "smolvlm",
+  "pad_token_id": 2,
+  "scale_factor": 2,
+  "text_config": {
+    "attention_bias": false,
+    "attention_dropout": 0.0,
+    "bos_token_id": 1,
+    "eos_token_id": 49279,
+    "head_dim": 32,
+    "hidden_act": "silu",
+    "hidden_size": 64,
+    "image_token_id": 49190,
+    "initializer_range": 0.02,
+    "intermediate_size": 56,
+    "max_position_embeddings": 256,
+    "mlp_bias": false,
+    "model_type": "llama",
+    "num_attention_heads": 2,
+    "num_hidden_layers": 2,
+    "num_key_value_heads": 2,
+    "pad_token_id": 2,
+    "pretraining_tp": 1,
+    "rms_norm_eps": 1e-06,
+    "rope_parameters": {
+      "rope_theta": 10000.0,
+      "rope_type": "default"
+    },
+    "sliding_window": 32,
+    "tie_word_embeddings": false,
+    "use_cache": true,
+    "vocab_size": 49280
+  },
+  "tie_word_embeddings": false,
+  "transformers_version": "5.16.0.dev0",
+  "use_cache": false,
+  "vision_config": {
+    "attention_dropout": 0.1,
+    "dropout": 0.1,
+    "hidden_act": "gelu_pytorch_tanh",
+    "hidden_size": 32,
+    "image_size": 16,
+    "initializer_range": 0.02,
+    "intermediate_size": 32,
+    "layer_norm_eps": 1e-06,
+    "model_type": "smolvlm_vision",
+    "num_attention_heads": 4,
+    "num_channels": 3,
+    "num_hidden_layers": 2,
+    "patch_size": 4
+  },
+  "vocab_size": 49280
+}

generation_config.json

@@ -0,0 +1,10 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 49279,
+  "output_attentions": false,
+  "output_hidden_states": false,
+  "pad_token_id": 2,
+  "transformers_version": "5.16.0.dev0",
+  "use_cache": false
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dacb3d9a73db914b69d681746f7e06fe1a51a9f2dc6e224f34b5b48d71c72683
+size 25549864

preprocessor_config.json

@@ -0,0 +1,32 @@
+{
+  "crop_size": {
+    "longest_edge": 16,
+    "shortest_edge": 16
+  },
+  "do_convert_rgb": true,
+  "do_image_splitting": true,
+  "do_normalize": true,
+  "do_pad": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "image_processor_type": "SmolVLMImageProcessor",
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "max_image_size": {
+    "longest_edge": 364
+  },
+  "resample": 1,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "longest_edge": 16,
+    "shortest_edge": 16
+  }
+}

tokenizer_config.json

@@ -0,0 +1,27 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": "<|im_start|>",
+  "clean_up_tokenization_spaces": false,
+  "end_of_utterance_token": "<end_of_utterance>",
+  "eos_token": "<end_of_utterance>",
+  "fake_image_token": "<fake_token_around_image>",
+  "global_image_token": "<global-img>",
+  "image_token": "<image>",
+  "is_local": true,
+  "legacy": false,
+  "local_files_only": false,
+  "model_max_length": 16384,
+  "model_specific_special_tokens": {
+    "end_of_utterance_token": "<end_of_utterance>",
+    "fake_image_token": "<fake_token_around_image>",
+    "global_image_token": "<global-img>",
+    "image_token": "<image>"
+  },
+  "pad_token": "<|im_end|>",
+  "processor_class": "SmolVLMProcessor",
+  "tokenizer_class": "TokenizersBackend",
+  "truncation_side": "left",
+  "unk_token": "<|endoftext|>",
+  "vocab_size": 49152
+}

video_preprocessor_config.json

@@ -0,0 +1,33 @@
+{
+  "crop_size": {
+    "longest_edge": 16,
+    "shortest_edge": 16
+  },
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_pad": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "do_sample_frames": false,
+  "image_mean": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "max_image_size": {
+    "longest_edge": 364
+  },
+  "resample": 1,
+  "rescale_factor": 0.00392156862745098,
+  "return_metadata": false,
+  "size": {
+    "longest_edge": 16,
+    "shortest_edge": 16
+  },
+  "video_processor_type": "SmolVLMVideoProcessor"
+}