Upload folder using huggingface_hub

README.md

@@ -0,0 +1,28 @@
+---
+base_model:
+- HuggingFaceTB/SmolLM2-1.7B-Instruct
+- google/siglip-so400m-patch14-384
+datasets:
+- HuggingFaceM4/the_cauldron
+- HuggingFaceM4/Docmatix
+language:
+- en
+library_name: transformers
+license: apache-2.0
+pipeline_tag: image-text-to-text
+tags:
+- mlx
+---
+
+# mlx-community/SmolVLM-Instruct-bf16
+This model was converted to MLX format from [`HuggingFaceTB/SmolVLM-Instruct`]() using mlx-vlm version **0.1.2**.
+Refer to the [original model card](https://huggingface.co/HuggingFaceTB/SmolVLM-Instruct) for more details on the model.
+## Use with mlx
+
+```bash
+pip install -U mlx-vlm
+```
+
+```bash
+python -m mlx_vlm.generate --model mlx-community/SmolVLM-Instruct-bf16 --max-tokens 100 --temp 0.0
+```

added_tokens.json

@@ -0,0 +1,5 @@
+{
+  "<end_of_utterance>": 49154,
+  "<fake_token_around_image>": 49152,
+  "<image>": 49153
+}

chat_template.json

@@ -0,0 +1,3 @@
+{
+  "chat_template": "<|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>\n{% endfor %}{% if add_generation_prompt %}{{ 'Assistant:' }}{% endif %}"
+}

config.json

@@ -0,0 +1,256 @@
+{
+    "architectures": [
+        "Idefics3ForConditionalGeneration"
+    ],
+    "image_seq_len": 81,
+    "image_token_id": 49153,
+    "model_type": "idefics3",
+    "scale_factor": 3,
+    "text_config": {
+        "_attn_implementation_autoset": false,
+        "_flash_attn_2_enabled": true,
+        "_name_or_path": "/fsx/m4/experiments/local_experiment_dir/s3_async_temporary_checkpoint_folder/tr_324_opt_400/unwrapped_model",
+        "add_cross_attention": false,
+        "architectures": [
+            "VLlama3ForCausalLM"
+        ],
+        "attention_bias": false,
+        "attention_dropout": 0.0,
+        "bad_words_ids": null,
+        "begin_suppress_tokens": null,
+        "bos_token_id": 0,
+        "chunk_size_feed_forward": 0,
+        "cross_attention_hidden_size": null,
+        "decoder_start_token_id": null,
+        "diversity_penalty": 0.0,
+        "do_sample": false,
+        "early_stopping": false,
+        "encoder_no_repeat_ngram_size": 0,
+        "eos_token_id": 0,
+        "exponential_decay_length_penalty": null,
+        "finetuning_task": null,
+        "forced_bos_token_id": null,
+        "forced_eos_token_id": null,
+        "head_dim": 64,
+        "hidden_act": "silu",
+        "hidden_size": 2048,
+        "id2label": {
+            "0": "LABEL_0",
+            "1": "LABEL_1"
+        },
+        "initializer_range": 0.02,
+        "intermediate_size": 8192,
+        "is_decoder": false,
+        "is_encoder_decoder": false,
+        "label2id": {
+            "LABEL_0": 0,
+            "LABEL_1": 1
+        },
+        "length_penalty": 1.0,
+        "max_length": 20,
+        "max_position_embeddings": 16384,
+        "min_length": 0,
+        "mlp_bias": false,
+        "model_type": "llama",
+        "neftune_noise_alpha": 0.0,
+        "no_repeat_ngram_size": 0,
+        "num_attention_heads": 32,
+        "num_beam_groups": 1,
+        "num_beams": 1,
+        "num_hidden_layers": 24,
+        "num_key_value_heads": 32,
+        "num_return_sequences": 1,
+        "output_attentions": false,
+        "output_hidden_states": false,
+        "output_scores": false,
+        "pad_token_id": 2,
+        "perceiver_config": {
+            "_attn_implementation_autoset": false,
+            "_name_or_path": "",
+            "add_cross_attention": false,
+            "architectures": null,
+            "attention_dropout": 0.0,
+            "bad_words_ids": null,
+            "begin_suppress_tokens": null,
+            "bos_token_id": null,
+            "chunk_size_feed_forward": 0,
+            "cross_attention_hidden_size": null,
+            "decoder_start_token_id": null,
+            "diversity_penalty": 0.0,
+            "do_sample": false,
+            "early_stopping": false,
+            "encoder_no_repeat_ngram_size": 0,
+            "eos_token_id": null,
+            "exponential_decay_length_penalty": null,
+            "finetuning_task": null,
+            "forced_bos_token_id": null,
+            "forced_eos_token_id": null,
+            "hidden_act": "silu",
+            "id2label": {
+                "0": "LABEL_0",
+                "1": "LABEL_1"
+            },
+            "is_decoder": false,
+            "is_encoder_decoder": false,
+            "label2id": {
+                "LABEL_0": 0,
+                "LABEL_1": 1
+            },
+            "length_penalty": 1.0,
+            "max_length": 20,
+            "min_length": 0,
+            "model_type": "vllama3",
+            "no_repeat_ngram_size": 0,
+            "num_beam_groups": 1,
+            "num_beams": 1,
+            "num_key_value_heads": 1,
+            "num_return_sequences": 1,
+            "output_attentions": false,
+            "output_hidden_states": false,
+            "output_scores": false,
+            "pad_token_id": null,
+            "prefix": null,
+            "problem_type": null,
+            "pruned_heads": {},
+            "qk_layer_norms_perceiver": false,
+            "remove_invalid_values": false,
+            "repetition_penalty": 1.0,
+            "resampler_depth": 6,
+            "resampler_head_dim": 96,
+            "resampler_n_heads": 16,
+            "resampler_n_latents": 64,
+            "return_dict": true,
+            "return_dict_in_generate": false,
+            "sep_token_id": null,
+            "suppress_tokens": null,
+            "task_specific_params": null,
+            "temperature": 1.0,
+            "tf_legacy_loss": false,
+            "tie_encoder_decoder": false,
+            "tie_word_embeddings": true,
+            "tokenizer_class": null,
+            "top_k": 50,
+            "top_p": 1.0,
+            "torch_dtype": null,
+            "torchscript": false,
+            "transformers_version": "4.46.0",
+            "typical_p": 1.0,
+            "use_bfloat16": false
+        },
+        "prefix": null,
+        "pretraining_tp": 1,
+        "problem_type": null,
+        "pruned_heads": {},
+        "qk_layer_norms": false,
+        "remove_invalid_values": false,
+        "repetition_penalty": 1.0,
+        "return_dict": true,
+        "return_dict_in_generate": false,
+        "rms_norm_eps": 1e-05,
+        "rope_scaling": null,
+        "rope_theta": 273768.0,
+        "sep_token_id": null,
+        "suppress_tokens": null,
+        "task_specific_params": null,
+        "temperature": 1.0,
+        "tf_legacy_loss": false,
+        "tie_encoder_decoder": false,
+        "tie_word_embeddings": false,
+        "tokenizer_class": null,
+        "top_k": 50,
+        "top_p": 1.0,
+        "torch_dtype": "bfloat16",
+        "torchscript": false,
+        "typical_p": 1.0,
+        "use_bfloat16": false,
+        "use_cache": true,
+        "use_resampler": false,
+        "vocab_size": 49155
+    },
+    "tie_word_embeddings": false,
+    "torch_dtype": "bfloat16",
+    "transformers_version": "4.46.0",
+    "use_cache": true,
+    "vision_config": {
+        "size": {
+            "longest_edge": 1920
+        },
+        "max_image_size": {
+            "longest_edge": 384
+        },
+        "_attn_implementation_autoset": false,
+        "_name_or_path": "",
+        "add_cross_attention": false,
+        "architectures": null,
+        "attention_dropout": 0.0,
+        "bad_words_ids": null,
+        "begin_suppress_tokens": null,
+        "bos_token_id": null,
+        "chunk_size_feed_forward": 0,
+        "cross_attention_hidden_size": null,
+        "decoder_start_token_id": null,
+        "diversity_penalty": 0.0,
+        "do_sample": false,
+        "early_stopping": false,
+        "encoder_no_repeat_ngram_size": 0,
+        "eos_token_id": null,
+        "exponential_decay_length_penalty": null,
+        "finetuning_task": null,
+        "forced_bos_token_id": null,
+        "forced_eos_token_id": null,
+        "hidden_act": "gelu_pytorch_tanh",
+        "hidden_size": 1152,
+        "id2label": {
+            "0": "LABEL_0",
+            "1": "LABEL_1"
+        },
+        "image_size": 384,
+        "initializer_range": 0.02,
+        "intermediate_size": 4304,
+        "is_decoder": false,
+        "is_encoder_decoder": false,
+        "label2id": {
+            "LABEL_0": 0,
+            "LABEL_1": 1
+        },
+        "layer_norm_eps": 1e-06,
+        "length_penalty": 1.0,
+        "max_length": 20,
+        "min_length": 0,
+        "model_type": "idefics3",
+        "no_repeat_ngram_size": 0,
+        "num_attention_heads": 16,
+        "num_beam_groups": 1,
+        "num_beams": 1,
+        "num_channels": 3,
+        "num_hidden_layers": 27,
+        "num_return_sequences": 1,
+        "output_attentions": false,
+        "output_hidden_states": false,
+        "output_scores": false,
+        "pad_token_id": null,
+        "patch_size": 14,
+        "prefix": null,
+        "problem_type": null,
+        "pruned_heads": {},
+        "remove_invalid_values": false,
+        "repetition_penalty": 1.0,
+        "return_dict": true,
+        "return_dict_in_generate": false,
+        "sep_token_id": null,
+        "suppress_tokens": null,
+        "task_specific_params": null,
+        "temperature": 1.0,
+        "tf_legacy_loss": false,
+        "tie_encoder_decoder": false,
+        "tie_word_embeddings": false,
+        "tokenizer_class": null,
+        "top_k": 50,
+        "top_p": 1.0,
+        "torch_dtype": null,
+        "torchscript": false,
+        "typical_p": 1.0,
+        "use_bfloat16": false
+    },
+    "vocab_size": 49155
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e15a96b68e5530d37dd15a7bd6c176d3e3002de9eacdede2cae2c3aa71fad750
+size 4492627795

model.safetensors.index.json

@@ -0,0 +1,664 @@
+{
+    "metadata": {
+        "total_size": 4492545760
+    },
+    "weight_map": {
+        "connector.modality_projection.proj.weight": "model.safetensors",
+        "language_model.embed_tokens.weight": "model.safetensors",
+        "language_model.layers.0.input_layernorm.weight": "model.safetensors",
+        "language_model.layers.0.mlp.down_proj.weight": "model.safetensors",
+        "language_model.layers.0.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.layers.0.mlp.up_proj.weight": "model.safetensors",
+        "language_model.layers.0.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.layers.0.self_attn.k_proj.weight": "model.safetensors",
+        "language_model.layers.0.self_attn.o_proj.weight": "model.safetensors",
+        "language_model.layers.0.self_attn.q_proj.weight": "model.safetensors",
+        "language_model.layers.0.self_attn.v_proj.weight": "model.safetensors",
+        "language_model.layers.1.input_layernorm.weight": "model.safetensors",
+        "language_model.layers.1.mlp.down_proj.weight": "model.safetensors",
+        "language_model.layers.1.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.layers.1.mlp.up_proj.weight": "model.safetensors",
+        "language_model.layers.1.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.layers.1.self_attn.k_proj.weight": "model.safetensors",
+        "language_model.layers.1.self_attn.o_proj.weight": "model.safetensors",
+        "language_model.layers.1.self_attn.q_proj.weight": "model.safetensors",
+        "language_model.layers.1.self_attn.v_proj.weight": "model.safetensors",
+        "language_model.layers.10.input_layernorm.weight": "model.safetensors",
+        "language_model.layers.10.mlp.down_proj.weight": "model.safetensors",
+        "language_model.layers.10.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.layers.10.mlp.up_proj.weight": "model.safetensors",
+        "language_model.layers.10.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.layers.10.self_attn.k_proj.weight": "model.safetensors",
+        "language_model.layers.10.self_attn.o_proj.weight": "model.safetensors",
+        "language_model.layers.10.self_attn.q_proj.weight": "model.safetensors",
+        "language_model.layers.10.self_attn.v_proj.weight": "model.safetensors",
+        "language_model.layers.11.input_layernorm.weight": "model.safetensors",
+        "language_model.layers.11.mlp.down_proj.weight": "model.safetensors",
+        "language_model.layers.11.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.layers.11.mlp.up_proj.weight": "model.safetensors",
+        "language_model.layers.11.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.layers.11.self_attn.k_proj.weight": "model.safetensors",
+        "language_model.layers.11.self_attn.o_proj.weight": "model.safetensors",
+        "language_model.layers.11.self_attn.q_proj.weight": "model.safetensors",
+        "language_model.layers.11.self_attn.v_proj.weight": "model.safetensors",
+        "language_model.layers.12.input_layernorm.weight": "model.safetensors",
+        "language_model.layers.12.mlp.down_proj.weight": "model.safetensors",
+        "language_model.layers.12.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.layers.12.mlp.up_proj.weight": "model.safetensors",
+        "language_model.layers.12.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.layers.12.self_attn.k_proj.weight": "model.safetensors",
+        "language_model.layers.12.self_attn.o_proj.weight": "model.safetensors",
+        "language_model.layers.12.self_attn.q_proj.weight": "model.safetensors",
+        "language_model.layers.12.self_attn.v_proj.weight": "model.safetensors",
+        "language_model.layers.13.input_layernorm.weight": "model.safetensors",
+        "language_model.layers.13.mlp.down_proj.weight": "model.safetensors",
+        "language_model.layers.13.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.layers.13.mlp.up_proj.weight": "model.safetensors",
+        "language_model.layers.13.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.layers.13.self_attn.k_proj.weight": "model.safetensors",
+        "language_model.layers.13.self_attn.o_proj.weight": "model.safetensors",
+        "language_model.layers.13.self_attn.q_proj.weight": "model.safetensors",
+        "language_model.layers.13.self_attn.v_proj.weight": "model.safetensors",
+        "language_model.layers.14.input_layernorm.weight": "model.safetensors",
+        "language_model.layers.14.mlp.down_proj.weight": "model.safetensors",
+        "language_model.layers.14.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.layers.14.mlp.up_proj.weight": "model.safetensors",
+        "language_model.layers.14.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.layers.14.self_attn.k_proj.weight": "model.safetensors",
+        "language_model.layers.14.self_attn.o_proj.weight": "model.safetensors",
+        "language_model.layers.14.self_attn.q_proj.weight": "model.safetensors",
+        "language_model.layers.14.self_attn.v_proj.weight": "model.safetensors",
+        "language_model.layers.15.input_layernorm.weight": "model.safetensors",
+        "language_model.layers.15.mlp.down_proj.weight": "model.safetensors",
+        "language_model.layers.15.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.layers.15.mlp.up_proj.weight": "model.safetensors",
+        "language_model.layers.15.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.layers.15.self_attn.k_proj.weight": "model.safetensors",
+        "language_model.layers.15.self_attn.o_proj.weight": "model.safetensors",
+        "language_model.layers.15.self_attn.q_proj.weight": "model.safetensors",
+        "language_model.layers.15.self_attn.v_proj.weight": "model.safetensors",
+        "language_model.layers.16.input_layernorm.weight": "model.safetensors",
+        "language_model.layers.16.mlp.down_proj.weight": "model.safetensors",
+        "language_model.layers.16.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.layers.16.mlp.up_proj.weight": "model.safetensors",
+        "language_model.layers.16.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.layers.16.self_attn.k_proj.weight": "model.safetensors",
+        "language_model.layers.16.self_attn.o_proj.weight": "model.safetensors",
+        "language_model.layers.16.self_attn.q_proj.weight": "model.safetensors",
+        "language_model.layers.16.self_attn.v_proj.weight": "model.safetensors",
+        "language_model.layers.17.input_layernorm.weight": "model.safetensors",
+        "language_model.layers.17.mlp.down_proj.weight": "model.safetensors",
+        "language_model.layers.17.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.layers.17.mlp.up_proj.weight": "model.safetensors",
+        "language_model.layers.17.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.layers.17.self_attn.k_proj.weight": "model.safetensors",
+        "language_model.layers.17.self_attn.o_proj.weight": "model.safetensors",
+        "language_model.layers.17.self_attn.q_proj.weight": "model.safetensors",
+        "language_model.layers.17.self_attn.v_proj.weight": "model.safetensors",
+        "language_model.layers.18.input_layernorm.weight": "model.safetensors",
+        "language_model.layers.18.mlp.down_proj.weight": "model.safetensors",
+        "language_model.layers.18.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.layers.18.mlp.up_proj.weight": "model.safetensors",
+        "language_model.layers.18.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.layers.18.self_attn.k_proj.weight": "model.safetensors",
+        "language_model.layers.18.self_attn.o_proj.weight": "model.safetensors",
+        "language_model.layers.18.self_attn.q_proj.weight": "model.safetensors",
+        "language_model.layers.18.self_attn.v_proj.weight": "model.safetensors",
+        "language_model.layers.19.input_layernorm.weight": "model.safetensors",
+        "language_model.layers.19.mlp.down_proj.weight": "model.safetensors",
+        "language_model.layers.19.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.layers.19.mlp.up_proj.weight": "model.safetensors",
+        "language_model.layers.19.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.layers.19.self_attn.k_proj.weight": "model.safetensors",
+        "language_model.layers.19.self_attn.o_proj.weight": "model.safetensors",
+        "language_model.layers.19.self_attn.q_proj.weight": "model.safetensors",
+        "language_model.layers.19.self_attn.v_proj.weight": "model.safetensors",
+        "language_model.layers.2.input_layernorm.weight": "model.safetensors",
+        "language_model.layers.2.mlp.down_proj.weight": "model.safetensors",
+        "language_model.layers.2.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.layers.2.mlp.up_proj.weight": "model.safetensors",
+        "language_model.layers.2.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.layers.2.self_attn.k_proj.weight": "model.safetensors",
+        "language_model.layers.2.self_attn.o_proj.weight": "model.safetensors",
+        "language_model.layers.2.self_attn.q_proj.weight": "model.safetensors",
+        "language_model.layers.2.self_attn.v_proj.weight": "model.safetensors",
+        "language_model.layers.20.input_layernorm.weight": "model.safetensors",
+        "language_model.layers.20.mlp.down_proj.weight": "model.safetensors",
+        "language_model.layers.20.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.layers.20.mlp.up_proj.weight": "model.safetensors",
+        "language_model.layers.20.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.layers.20.self_attn.k_proj.weight": "model.safetensors",
+        "language_model.layers.20.self_attn.o_proj.weight": "model.safetensors",
+        "language_model.layers.20.self_attn.q_proj.weight": "model.safetensors",
+        "language_model.layers.20.self_attn.v_proj.weight": "model.safetensors",
+        "language_model.layers.21.input_layernorm.weight": "model.safetensors",
+        "language_model.layers.21.mlp.down_proj.weight": "model.safetensors",
+        "language_model.layers.21.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.layers.21.mlp.up_proj.weight": "model.safetensors",
+        "language_model.layers.21.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.layers.21.self_attn.k_proj.weight": "model.safetensors",
+        "language_model.layers.21.self_attn.o_proj.weight": "model.safetensors",
+        "language_model.layers.21.self_attn.q_proj.weight": "model.safetensors",
+        "language_model.layers.21.self_attn.v_proj.weight": "model.safetensors",
+        "language_model.layers.22.input_layernorm.weight": "model.safetensors",
+        "language_model.layers.22.mlp.down_proj.weight": "model.safetensors",
+        "language_model.layers.22.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.layers.22.mlp.up_proj.weight": "model.safetensors",
+        "language_model.layers.22.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.layers.22.self_attn.k_proj.weight": "model.safetensors",
+        "language_model.layers.22.self_attn.o_proj.weight": "model.safetensors",
+        "language_model.layers.22.self_attn.q_proj.weight": "model.safetensors",
+        "language_model.layers.22.self_attn.v_proj.weight": "model.safetensors",
+        "language_model.layers.23.input_layernorm.weight": "model.safetensors",
+        "language_model.layers.23.mlp.down_proj.weight": "model.safetensors",
+        "language_model.layers.23.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.layers.23.mlp.up_proj.weight": "model.safetensors",
+        "language_model.layers.23.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.layers.23.self_attn.k_proj.weight": "model.safetensors",
+        "language_model.layers.23.self_attn.o_proj.weight": "model.safetensors",
+        "language_model.layers.23.self_attn.q_proj.weight": "model.safetensors",
+        "language_model.layers.23.self_attn.v_proj.weight": "model.safetensors",
+        "language_model.layers.3.input_layernorm.weight": "model.safetensors",
+        "language_model.layers.3.mlp.down_proj.weight": "model.safetensors",
+        "language_model.layers.3.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.layers.3.mlp.up_proj.weight": "model.safetensors",
+        "language_model.layers.3.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.layers.3.self_attn.k_proj.weight": "model.safetensors",
+        "language_model.layers.3.self_attn.o_proj.weight": "model.safetensors",
+        "language_model.layers.3.self_attn.q_proj.weight": "model.safetensors",
+        "language_model.layers.3.self_attn.v_proj.weight": "model.safetensors",
+        "language_model.layers.4.input_layernorm.weight": "model.safetensors",
+        "language_model.layers.4.mlp.down_proj.weight": "model.safetensors",
+        "language_model.layers.4.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.layers.4.mlp.up_proj.weight": "model.safetensors",
+        "language_model.layers.4.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.layers.4.self_attn.k_proj.weight": "model.safetensors",
+        "language_model.layers.4.self_attn.o_proj.weight": "model.safetensors",
+        "language_model.layers.4.self_attn.q_proj.weight": "model.safetensors",
+        "language_model.layers.4.self_attn.v_proj.weight": "model.safetensors",
+        "language_model.layers.5.input_layernorm.weight": "model.safetensors",
+        "language_model.layers.5.mlp.down_proj.weight": "model.safetensors",
+        "language_model.layers.5.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.layers.5.mlp.up_proj.weight": "model.safetensors",
+        "language_model.layers.5.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.layers.5.self_attn.k_proj.weight": "model.safetensors",
+        "language_model.layers.5.self_attn.o_proj.weight": "model.safetensors",
+        "language_model.layers.5.self_attn.q_proj.weight": "model.safetensors",
+        "language_model.layers.5.self_attn.v_proj.weight": "model.safetensors",
+        "language_model.layers.6.input_layernorm.weight": "model.safetensors",
+        "language_model.layers.6.mlp.down_proj.weight": "model.safetensors",
+        "language_model.layers.6.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.layers.6.mlp.up_proj.weight": "model.safetensors",
+        "language_model.layers.6.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.layers.6.self_attn.k_proj.weight": "model.safetensors",
+        "language_model.layers.6.self_attn.o_proj.weight": "model.safetensors",
+        "language_model.layers.6.self_attn.q_proj.weight": "model.safetensors",
+        "language_model.layers.6.self_attn.v_proj.weight": "model.safetensors",
+        "language_model.layers.7.input_layernorm.weight": "model.safetensors",
+        "language_model.layers.7.mlp.down_proj.weight": "model.safetensors",
+        "language_model.layers.7.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.layers.7.mlp.up_proj.weight": "model.safetensors",
+        "language_model.layers.7.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.layers.7.self_attn.k_proj.weight": "model.safetensors",
+        "language_model.layers.7.self_attn.o_proj.weight": "model.safetensors",
+        "language_model.layers.7.self_attn.q_proj.weight": "model.safetensors",
+        "language_model.layers.7.self_attn.v_proj.weight": "model.safetensors",
+        "language_model.layers.8.input_layernorm.weight": "model.safetensors",
+        "language_model.layers.8.mlp.down_proj.weight": "model.safetensors",
+        "language_model.layers.8.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.layers.8.mlp.up_proj.weight": "model.safetensors",
+        "language_model.layers.8.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.layers.8.self_attn.k_proj.weight": "model.safetensors",
+        "language_model.layers.8.self_attn.o_proj.weight": "model.safetensors",
+        "language_model.layers.8.self_attn.q_proj.weight": "model.safetensors",
+        "language_model.layers.8.self_attn.v_proj.weight": "model.safetensors",
+        "language_model.layers.9.input_layernorm.weight": "model.safetensors",
+        "language_model.layers.9.mlp.down_proj.weight": "model.safetensors",
+        "language_model.layers.9.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.layers.9.mlp.up_proj.weight": "model.safetensors",
+        "language_model.layers.9.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.layers.9.self_attn.k_proj.weight": "model.safetensors",
+        "language_model.layers.9.self_attn.o_proj.weight": "model.safetensors",
+        "language_model.layers.9.self_attn.q_proj.weight": "model.safetensors",
+        "language_model.layers.9.self_attn.v_proj.weight": "model.safetensors",
+        "language_model.lm_head.weight": "model.safetensors",
+        "language_model.norm.weight": "model.safetensors",
+        "vision_model.embeddings.patch_embedding.bias": "model.safetensors",
+        "vision_model.embeddings.patch_embedding.weight": "model.safetensors",
+        "vision_model.embeddings.position_embedding.weight": "model.safetensors",
+        "vision_model.encoder.layers.0.layer_norm1.bias": "model.safetensors",
+        "vision_model.encoder.layers.0.layer_norm1.weight": "model.safetensors",
+        "vision_model.encoder.layers.0.layer_norm2.bias": "model.safetensors",
+        "vision_model.encoder.layers.0.layer_norm2.weight": "model.safetensors",
+        "vision_model.encoder.layers.0.mlp.fc1.bias": "model.safetensors",
+        "vision_model.encoder.layers.0.mlp.fc1.weight": "model.safetensors",
+        "vision_model.encoder.layers.0.mlp.fc2.bias": "model.safetensors",
+        "vision_model.encoder.layers.0.mlp.fc2.weight": "model.safetensors",
+        "vision_model.encoder.layers.0.self_attn.k_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.0.self_attn.k_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.0.self_attn.out_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.0.self_attn.out_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.0.self_attn.q_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.0.self_attn.q_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.0.self_attn.v_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.0.self_attn.v_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.1.layer_norm1.bias": "model.safetensors",
+        "vision_model.encoder.layers.1.layer_norm1.weight": "model.safetensors",
+        "vision_model.encoder.layers.1.layer_norm2.bias": "model.safetensors",
+        "vision_model.encoder.layers.1.layer_norm2.weight": "model.safetensors",
+        "vision_model.encoder.layers.1.mlp.fc1.bias": "model.safetensors",
+        "vision_model.encoder.layers.1.mlp.fc1.weight": "model.safetensors",
+        "vision_model.encoder.layers.1.mlp.fc2.bias": "model.safetensors",
+        "vision_model.encoder.layers.1.mlp.fc2.weight": "model.safetensors",
+        "vision_model.encoder.layers.1.self_attn.k_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.1.self_attn.k_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.1.self_attn.out_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.1.self_attn.out_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.1.self_attn.q_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.1.self_attn.q_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.1.self_attn.v_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.1.self_attn.v_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.10.layer_norm1.bias": "model.safetensors",
+        "vision_model.encoder.layers.10.layer_norm1.weight": "model.safetensors",
+        "vision_model.encoder.layers.10.layer_norm2.bias": "model.safetensors",
+        "vision_model.encoder.layers.10.layer_norm2.weight": "model.safetensors",
+        "vision_model.encoder.layers.10.mlp.fc1.bias": "model.safetensors",
+        "vision_model.encoder.layers.10.mlp.fc1.weight": "model.safetensors",
+        "vision_model.encoder.layers.10.mlp.fc2.bias": "model.safetensors",
+        "vision_model.encoder.layers.10.mlp.fc2.weight": "model.safetensors",
+        "vision_model.encoder.layers.10.self_attn.k_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.10.self_attn.k_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.10.self_attn.out_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.10.self_attn.out_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.10.self_attn.q_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.10.self_attn.q_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.10.self_attn.v_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.10.self_attn.v_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.11.layer_norm1.bias": "model.safetensors",
+        "vision_model.encoder.layers.11.layer_norm1.weight": "model.safetensors",
+        "vision_model.encoder.layers.11.layer_norm2.bias": "model.safetensors",
+        "vision_model.encoder.layers.11.layer_norm2.weight": "model.safetensors",
+        "vision_model.encoder.layers.11.mlp.fc1.bias": "model.safetensors",
+        "vision_model.encoder.layers.11.mlp.fc1.weight": "model.safetensors",
+        "vision_model.encoder.layers.11.mlp.fc2.bias": "model.safetensors",
+        "vision_model.encoder.layers.11.mlp.fc2.weight": "model.safetensors",
+        "vision_model.encoder.layers.11.self_attn.k_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.11.self_attn.k_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.11.self_attn.out_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.11.self_attn.out_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.11.self_attn.q_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.11.self_attn.q_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.11.self_attn.v_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.11.self_attn.v_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.12.layer_norm1.bias": "model.safetensors",
+        "vision_model.encoder.layers.12.layer_norm1.weight": "model.safetensors",
+        "vision_model.encoder.layers.12.layer_norm2.bias": "model.safetensors",
+        "vision_model.encoder.layers.12.layer_norm2.weight": "model.safetensors",
+        "vision_model.encoder.layers.12.mlp.fc1.bias": "model.safetensors",
+        "vision_model.encoder.layers.12.mlp.fc1.weight": "model.safetensors",
+        "vision_model.encoder.layers.12.mlp.fc2.bias": "model.safetensors",
+        "vision_model.encoder.layers.12.mlp.fc2.weight": "model.safetensors",
+        "vision_model.encoder.layers.12.self_attn.k_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.12.self_attn.k_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.12.self_attn.out_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.12.self_attn.out_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.12.self_attn.q_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.12.self_attn.q_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.12.self_attn.v_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.12.self_attn.v_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.13.layer_norm1.bias": "model.safetensors",
+        "vision_model.encoder.layers.13.layer_norm1.weight": "model.safetensors",
+        "vision_model.encoder.layers.13.layer_norm2.bias": "model.safetensors",
+        "vision_model.encoder.layers.13.layer_norm2.weight": "model.safetensors",
+        "vision_model.encoder.layers.13.mlp.fc1.bias": "model.safetensors",
+        "vision_model.encoder.layers.13.mlp.fc1.weight": "model.safetensors",
+        "vision_model.encoder.layers.13.mlp.fc2.bias": "model.safetensors",
+        "vision_model.encoder.layers.13.mlp.fc2.weight": "model.safetensors",
+        "vision_model.encoder.layers.13.self_attn.k_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.13.self_attn.k_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.13.self_attn.out_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.13.self_attn.out_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.13.self_attn.q_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.13.self_attn.q_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.13.self_attn.v_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.13.self_attn.v_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.14.layer_norm1.bias": "model.safetensors",
+        "vision_model.encoder.layers.14.layer_norm1.weight": "model.safetensors",
+        "vision_model.encoder.layers.14.layer_norm2.bias": "model.safetensors",
+        "vision_model.encoder.layers.14.layer_norm2.weight": "model.safetensors",
+        "vision_model.encoder.layers.14.mlp.fc1.bias": "model.safetensors",
+        "vision_model.encoder.layers.14.mlp.fc1.weight": "model.safetensors",
+        "vision_model.encoder.layers.14.mlp.fc2.bias": "model.safetensors",
+        "vision_model.encoder.layers.14.mlp.fc2.weight": "model.safetensors",
+        "vision_model.encoder.layers.14.self_attn.k_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.14.self_attn.k_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.14.self_attn.out_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.14.self_attn.out_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.14.self_attn.q_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.14.self_attn.q_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.14.self_attn.v_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.14.self_attn.v_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.15.layer_norm1.bias": "model.safetensors",
+        "vision_model.encoder.layers.15.layer_norm1.weight": "model.safetensors",
+        "vision_model.encoder.layers.15.layer_norm2.bias": "model.safetensors",
+        "vision_model.encoder.layers.15.layer_norm2.weight": "model.safetensors",
+        "vision_model.encoder.layers.15.mlp.fc1.bias": "model.safetensors",
+        "vision_model.encoder.layers.15.mlp.fc1.weight": "model.safetensors",
+        "vision_model.encoder.layers.15.mlp.fc2.bias": "model.safetensors",
+        "vision_model.encoder.layers.15.mlp.fc2.weight": "model.safetensors",
+        "vision_model.encoder.layers.15.self_attn.k_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.15.self_attn.k_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.15.self_attn.out_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.15.self_attn.out_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.15.self_attn.q_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.15.self_attn.q_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.15.self_attn.v_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.15.self_attn.v_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.16.layer_norm1.bias": "model.safetensors",
+        "vision_model.encoder.layers.16.layer_norm1.weight": "model.safetensors",
+        "vision_model.encoder.layers.16.layer_norm2.bias": "model.safetensors",
+        "vision_model.encoder.layers.16.layer_norm2.weight": "model.safetensors",
+        "vision_model.encoder.layers.16.mlp.fc1.bias": "model.safetensors",
+        "vision_model.encoder.layers.16.mlp.fc1.weight": "model.safetensors",
+        "vision_model.encoder.layers.16.mlp.fc2.bias": "model.safetensors",
+        "vision_model.encoder.layers.16.mlp.fc2.weight": "model.safetensors",
+        "vision_model.encoder.layers.16.self_attn.k_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.16.self_attn.k_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.16.self_attn.out_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.16.self_attn.out_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.16.self_attn.q_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.16.self_attn.q_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.16.self_attn.v_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.16.self_attn.v_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.17.layer_norm1.bias": "model.safetensors",
+        "vision_model.encoder.layers.17.layer_norm1.weight": "model.safetensors",
+        "vision_model.encoder.layers.17.layer_norm2.bias": "model.safetensors",
+        "vision_model.encoder.layers.17.layer_norm2.weight": "model.safetensors",
+        "vision_model.encoder.layers.17.mlp.fc1.bias": "model.safetensors",
+        "vision_model.encoder.layers.17.mlp.fc1.weight": "model.safetensors",
+        "vision_model.encoder.layers.17.mlp.fc2.bias": "model.safetensors",
+        "vision_model.encoder.layers.17.mlp.fc2.weight": "model.safetensors",
+        "vision_model.encoder.layers.17.self_attn.k_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.17.self_attn.k_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.17.self_attn.out_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.17.self_attn.out_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.17.self_attn.q_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.17.self_attn.q_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.17.self_attn.v_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.17.self_attn.v_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.18.layer_norm1.bias": "model.safetensors",
+        "vision_model.encoder.layers.18.layer_norm1.weight": "model.safetensors",
+        "vision_model.encoder.layers.18.layer_norm2.bias": "model.safetensors",
+        "vision_model.encoder.layers.18.layer_norm2.weight": "model.safetensors",
+        "vision_model.encoder.layers.18.mlp.fc1.bias": "model.safetensors",
+        "vision_model.encoder.layers.18.mlp.fc1.weight": "model.safetensors",
+        "vision_model.encoder.layers.18.mlp.fc2.bias": "model.safetensors",
+        "vision_model.encoder.layers.18.mlp.fc2.weight": "model.safetensors",
+        "vision_model.encoder.layers.18.self_attn.k_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.18.self_attn.k_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.18.self_attn.out_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.18.self_attn.out_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.18.self_attn.q_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.18.self_attn.q_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.18.self_attn.v_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.18.self_attn.v_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.19.layer_norm1.bias": "model.safetensors",
+        "vision_model.encoder.layers.19.layer_norm1.weight": "model.safetensors",
+        "vision_model.encoder.layers.19.layer_norm2.bias": "model.safetensors",
+        "vision_model.encoder.layers.19.layer_norm2.weight": "model.safetensors",
+        "vision_model.encoder.layers.19.mlp.fc1.bias": "model.safetensors",
+        "vision_model.encoder.layers.19.mlp.fc1.weight": "model.safetensors",
+        "vision_model.encoder.layers.19.mlp.fc2.bias": "model.safetensors",
+        "vision_model.encoder.layers.19.mlp.fc2.weight": "model.safetensors",
+        "vision_model.encoder.layers.19.self_attn.k_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.19.self_attn.k_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.19.self_attn.out_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.19.self_attn.out_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.19.self_attn.q_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.19.self_attn.q_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.19.self_attn.v_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.19.self_attn.v_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.2.layer_norm1.bias": "model.safetensors",
+        "vision_model.encoder.layers.2.layer_norm1.weight": "model.safetensors",
+        "vision_model.encoder.layers.2.layer_norm2.bias": "model.safetensors",
+        "vision_model.encoder.layers.2.layer_norm2.weight": "model.safetensors",
+        "vision_model.encoder.layers.2.mlp.fc1.bias": "model.safetensors",
+        "vision_model.encoder.layers.2.mlp.fc1.weight": "model.safetensors",
+        "vision_model.encoder.layers.2.mlp.fc2.bias": "model.safetensors",
+        "vision_model.encoder.layers.2.mlp.fc2.weight": "model.safetensors",
+        "vision_model.encoder.layers.2.self_attn.k_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.2.self_attn.k_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.2.self_attn.out_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.2.self_attn.out_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.2.self_attn.q_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.2.self_attn.q_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.2.self_attn.v_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.2.self_attn.v_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.20.layer_norm1.bias": "model.safetensors",
+        "vision_model.encoder.layers.20.layer_norm1.weight": "model.safetensors",
+        "vision_model.encoder.layers.20.layer_norm2.bias": "model.safetensors",
+        "vision_model.encoder.layers.20.layer_norm2.weight": "model.safetensors",
+        "vision_model.encoder.layers.20.mlp.fc1.bias": "model.safetensors",
+        "vision_model.encoder.layers.20.mlp.fc1.weight": "model.safetensors",
+        "vision_model.encoder.layers.20.mlp.fc2.bias": "model.safetensors",
+        "vision_model.encoder.layers.20.mlp.fc2.weight": "model.safetensors",
+        "vision_model.encoder.layers.20.self_attn.k_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.20.self_attn.k_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.20.self_attn.out_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.20.self_attn.out_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.20.self_attn.q_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.20.self_attn.q_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.20.self_attn.v_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.20.self_attn.v_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.21.layer_norm1.bias": "model.safetensors",
+        "vision_model.encoder.layers.21.layer_norm1.weight": "model.safetensors",
+        "vision_model.encoder.layers.21.layer_norm2.bias": "model.safetensors",
+        "vision_model.encoder.layers.21.layer_norm2.weight": "model.safetensors",
+        "vision_model.encoder.layers.21.mlp.fc1.bias": "model.safetensors",
+        "vision_model.encoder.layers.21.mlp.fc1.weight": "model.safetensors",
+        "vision_model.encoder.layers.21.mlp.fc2.bias": "model.safetensors",
+        "vision_model.encoder.layers.21.mlp.fc2.weight": "model.safetensors",
+        "vision_model.encoder.layers.21.self_attn.k_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.21.self_attn.k_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.21.self_attn.out_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.21.self_attn.out_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.21.self_attn.q_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.21.self_attn.q_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.21.self_attn.v_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.21.self_attn.v_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.22.layer_norm1.bias": "model.safetensors",
+        "vision_model.encoder.layers.22.layer_norm1.weight": "model.safetensors",
+        "vision_model.encoder.layers.22.layer_norm2.bias": "model.safetensors",
+        "vision_model.encoder.layers.22.layer_norm2.weight": "model.safetensors",
+        "vision_model.encoder.layers.22.mlp.fc1.bias": "model.safetensors",
+        "vision_model.encoder.layers.22.mlp.fc1.weight": "model.safetensors",
+        "vision_model.encoder.layers.22.mlp.fc2.bias": "model.safetensors",
+        "vision_model.encoder.layers.22.mlp.fc2.weight": "model.safetensors",
+        "vision_model.encoder.layers.22.self_attn.k_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.22.self_attn.k_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.22.self_attn.out_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.22.self_attn.out_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.22.self_attn.q_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.22.self_attn.q_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.22.self_attn.v_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.22.self_attn.v_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.23.layer_norm1.bias": "model.safetensors",
+        "vision_model.encoder.layers.23.layer_norm1.weight": "model.safetensors",
+        "vision_model.encoder.layers.23.layer_norm2.bias": "model.safetensors",
+        "vision_model.encoder.layers.23.layer_norm2.weight": "model.safetensors",
+        "vision_model.encoder.layers.23.mlp.fc1.bias": "model.safetensors",
+        "vision_model.encoder.layers.23.mlp.fc1.weight": "model.safetensors",
+        "vision_model.encoder.layers.23.mlp.fc2.bias": "model.safetensors",
+        "vision_model.encoder.layers.23.mlp.fc2.weight": "model.safetensors",
+        "vision_model.encoder.layers.23.self_attn.k_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.23.self_attn.k_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.23.self_attn.out_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.23.self_attn.out_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.23.self_attn.q_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.23.self_attn.q_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.23.self_attn.v_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.23.self_attn.v_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.24.layer_norm1.bias": "model.safetensors",
+        "vision_model.encoder.layers.24.layer_norm1.weight": "model.safetensors",
+        "vision_model.encoder.layers.24.layer_norm2.bias": "model.safetensors",
+        "vision_model.encoder.layers.24.layer_norm2.weight": "model.safetensors",
+        "vision_model.encoder.layers.24.mlp.fc1.bias": "model.safetensors",
+        "vision_model.encoder.layers.24.mlp.fc1.weight": "model.safetensors",
+        "vision_model.encoder.layers.24.mlp.fc2.bias": "model.safetensors",
+        "vision_model.encoder.layers.24.mlp.fc2.weight": "model.safetensors",
+        "vision_model.encoder.layers.24.self_attn.k_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.24.self_attn.k_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.24.self_attn.out_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.24.self_attn.out_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.24.self_attn.q_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.24.self_attn.q_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.24.self_attn.v_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.24.self_attn.v_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.25.layer_norm1.bias": "model.safetensors",
+        "vision_model.encoder.layers.25.layer_norm1.weight": "model.safetensors",
+        "vision_model.encoder.layers.25.layer_norm2.bias": "model.safetensors",
+        "vision_model.encoder.layers.25.layer_norm2.weight": "model.safetensors",
+        "vision_model.encoder.layers.25.mlp.fc1.bias": "model.safetensors",
+        "vision_model.encoder.layers.25.mlp.fc1.weight": "model.safetensors",
+        "vision_model.encoder.layers.25.mlp.fc2.bias": "model.safetensors",
+        "vision_model.encoder.layers.25.mlp.fc2.weight": "model.safetensors",
+        "vision_model.encoder.layers.25.self_attn.k_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.25.self_attn.k_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.25.self_attn.out_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.25.self_attn.out_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.25.self_attn.q_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.25.self_attn.q_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.25.self_attn.v_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.25.self_attn.v_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.26.layer_norm1.bias": "model.safetensors",
+        "vision_model.encoder.layers.26.layer_norm1.weight": "model.safetensors",
+        "vision_model.encoder.layers.26.layer_norm2.bias": "model.safetensors",
+        "vision_model.encoder.layers.26.layer_norm2.weight": "model.safetensors",
+        "vision_model.encoder.layers.26.mlp.fc1.bias": "model.safetensors",
+        "vision_model.encoder.layers.26.mlp.fc1.weight": "model.safetensors",
+        "vision_model.encoder.layers.26.mlp.fc2.bias": "model.safetensors",
+        "vision_model.encoder.layers.26.mlp.fc2.weight": "model.safetensors",
+        "vision_model.encoder.layers.26.self_attn.k_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.26.self_attn.k_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.26.self_attn.out_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.26.self_attn.out_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.26.self_attn.q_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.26.self_attn.q_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.26.self_attn.v_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.26.self_attn.v_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.3.layer_norm1.bias": "model.safetensors",
+        "vision_model.encoder.layers.3.layer_norm1.weight": "model.safetensors",
+        "vision_model.encoder.layers.3.layer_norm2.bias": "model.safetensors",
+        "vision_model.encoder.layers.3.layer_norm2.weight": "model.safetensors",
+        "vision_model.encoder.layers.3.mlp.fc1.bias": "model.safetensors",
+        "vision_model.encoder.layers.3.mlp.fc1.weight": "model.safetensors",
+        "vision_model.encoder.layers.3.mlp.fc2.bias": "model.safetensors",
+        "vision_model.encoder.layers.3.mlp.fc2.weight": "model.safetensors",
+        "vision_model.encoder.layers.3.self_attn.k_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.3.self_attn.k_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.3.self_attn.out_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.3.self_attn.out_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.3.self_attn.q_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.3.self_attn.q_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.3.self_attn.v_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.3.self_attn.v_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.4.layer_norm1.bias": "model.safetensors",
+        "vision_model.encoder.layers.4.layer_norm1.weight": "model.safetensors",
+        "vision_model.encoder.layers.4.layer_norm2.bias": "model.safetensors",
+        "vision_model.encoder.layers.4.layer_norm2.weight": "model.safetensors",
+        "vision_model.encoder.layers.4.mlp.fc1.bias": "model.safetensors",
+        "vision_model.encoder.layers.4.mlp.fc1.weight": "model.safetensors",
+        "vision_model.encoder.layers.4.mlp.fc2.bias": "model.safetensors",
+        "vision_model.encoder.layers.4.mlp.fc2.weight": "model.safetensors",
+        "vision_model.encoder.layers.4.self_attn.k_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.4.self_attn.k_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.4.self_attn.out_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.4.self_attn.out_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.4.self_attn.q_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.4.self_attn.q_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.4.self_attn.v_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.4.self_attn.v_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.5.layer_norm1.bias": "model.safetensors",
+        "vision_model.encoder.layers.5.layer_norm1.weight": "model.safetensors",
+        "vision_model.encoder.layers.5.layer_norm2.bias": "model.safetensors",
+        "vision_model.encoder.layers.5.layer_norm2.weight": "model.safetensors",
+        "vision_model.encoder.layers.5.mlp.fc1.bias": "model.safetensors",
+        "vision_model.encoder.layers.5.mlp.fc1.weight": "model.safetensors",
+        "vision_model.encoder.layers.5.mlp.fc2.bias": "model.safetensors",
+        "vision_model.encoder.layers.5.mlp.fc2.weight": "model.safetensors",
+        "vision_model.encoder.layers.5.self_attn.k_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.5.self_attn.k_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.5.self_attn.out_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.5.self_attn.out_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.5.self_attn.q_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.5.self_attn.q_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.5.self_attn.v_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.5.self_attn.v_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.6.layer_norm1.bias": "model.safetensors",
+        "vision_model.encoder.layers.6.layer_norm1.weight": "model.safetensors",
+        "vision_model.encoder.layers.6.layer_norm2.bias": "model.safetensors",
+        "vision_model.encoder.layers.6.layer_norm2.weight": "model.safetensors",
+        "vision_model.encoder.layers.6.mlp.fc1.bias": "model.safetensors",
+        "vision_model.encoder.layers.6.mlp.fc1.weight": "model.safetensors",
+        "vision_model.encoder.layers.6.mlp.fc2.bias": "model.safetensors",
+        "vision_model.encoder.layers.6.mlp.fc2.weight": "model.safetensors",
+        "vision_model.encoder.layers.6.self_attn.k_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.6.self_attn.k_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.6.self_attn.out_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.6.self_attn.out_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.6.self_attn.q_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.6.self_attn.q_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.6.self_attn.v_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.6.self_attn.v_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.7.layer_norm1.bias": "model.safetensors",
+        "vision_model.encoder.layers.7.layer_norm1.weight": "model.safetensors",
+        "vision_model.encoder.layers.7.layer_norm2.bias": "model.safetensors",
+        "vision_model.encoder.layers.7.layer_norm2.weight": "model.safetensors",
+        "vision_model.encoder.layers.7.mlp.fc1.bias": "model.safetensors",
+        "vision_model.encoder.layers.7.mlp.fc1.weight": "model.safetensors",
+        "vision_model.encoder.layers.7.mlp.fc2.bias": "model.safetensors",
+        "vision_model.encoder.layers.7.mlp.fc2.weight": "model.safetensors",
+        "vision_model.encoder.layers.7.self_attn.k_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.7.self_attn.k_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.7.self_attn.out_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.7.self_attn.out_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.7.self_attn.q_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.7.self_attn.q_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.7.self_attn.v_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.7.self_attn.v_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.8.layer_norm1.bias": "model.safetensors",
+        "vision_model.encoder.layers.8.layer_norm1.weight": "model.safetensors",
+        "vision_model.encoder.layers.8.layer_norm2.bias": "model.safetensors",
+        "vision_model.encoder.layers.8.layer_norm2.weight": "model.safetensors",
+        "vision_model.encoder.layers.8.mlp.fc1.bias": "model.safetensors",
+        "vision_model.encoder.layers.8.mlp.fc1.weight": "model.safetensors",
+        "vision_model.encoder.layers.8.mlp.fc2.bias": "model.safetensors",
+        "vision_model.encoder.layers.8.mlp.fc2.weight": "model.safetensors",
+        "vision_model.encoder.layers.8.self_attn.k_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.8.self_attn.k_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.8.self_attn.out_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.8.self_attn.out_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.8.self_attn.q_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.8.self_attn.q_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.8.self_attn.v_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.8.self_attn.v_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.9.layer_norm1.bias": "model.safetensors",
+        "vision_model.encoder.layers.9.layer_norm1.weight": "model.safetensors",
+        "vision_model.encoder.layers.9.layer_norm2.bias": "model.safetensors",
+        "vision_model.encoder.layers.9.layer_norm2.weight": "model.safetensors",
+        "vision_model.encoder.layers.9.mlp.fc1.bias": "model.safetensors",
+        "vision_model.encoder.layers.9.mlp.fc1.weight": "model.safetensors",
+        "vision_model.encoder.layers.9.mlp.fc2.bias": "model.safetensors",
+        "vision_model.encoder.layers.9.mlp.fc2.weight": "model.safetensors",
+        "vision_model.encoder.layers.9.self_attn.k_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.9.self_attn.k_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.9.self_attn.out_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.9.self_attn.out_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.9.self_attn.q_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.9.self_attn.q_proj.weight": "model.safetensors",
+        "vision_model.encoder.layers.9.self_attn.v_proj.bias": "model.safetensors",
+        "vision_model.encoder.layers.9.self_attn.v_proj.weight": "model.safetensors",
+        "vision_model.post_layernorm.bias": "model.safetensors",
+        "vision_model.post_layernorm.weight": "model.safetensors"
+    }
+}

preprocessor_config.json

@@ -0,0 +1,28 @@
+{
+  "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": "Idefics3ImageProcessor",
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "max_image_size": {
+    "longest_edge": 384
+  },
+  "processor_class": "Idefics3Processor",
+  "resample": 1,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "longest_edge": 1536
+  }
+}

processor_config.json

@@ -0,0 +1,4 @@
+{
+  "image_seq_len": 81,
+  "processor_class": "Idefics3Processor"
+}

special_tokens_map.json

@@ -0,0 +1,53 @@
+{
+  "additional_special_tokens": [
+    {
+      "content": "<fake_token_around_image>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false
+    },
+    {
+      "content": "<image>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false
+    },
+    {
+      "content": "<end_of_utterance>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false
+    }
+  ],
+  "bos_token": {
+    "content": "<|im_start|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<end_of_utterance>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,182 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<repo_name>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<reponame>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "5": {
+      "content": "<file_sep>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "6": {
+      "content": "<filename>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "7": {
+      "content": "<gh_stars>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "8": {
+      "content": "<issue_start>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "9": {
+      "content": "<issue_comment>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "10": {
+      "content": "<issue_closed>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "11": {
+      "content": "<jupyter_start>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "12": {
+      "content": "<jupyter_text>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "13": {
+      "content": "<jupyter_code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "14": {
+      "content": "<jupyter_output>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "15": {
+      "content": "<jupyter_script>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "16": {
+      "content": "<empty_output>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "49152": {
+      "content": "<fake_token_around_image>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "49153": {
+      "content": "<image>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "49154": {
+      "content": "<end_of_utterance>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<fake_token_around_image>",
+    "<image>",
+    "<end_of_utterance>"
+  ],
+  "bos_token": "<|im_start|>",
+  "chat_template": "{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<end_of_utterance>",
+  "legacy": false,
+  "model_max_length": 16384,
+  "pad_token": "<|im_end|>",
+  "processor_class": "Idefics3Processor",
+  "tokenizer_class": "GPT2Tokenizer",
+  "truncation_side": "left",
+  "unk_token": "<|endoftext|>",
+  "vocab_size": 49152
+}