Upload folder using huggingface_hub

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 +1,340 @@
+{
+  "_name_or_path": "HuggingFaceTB/SmolVLM-Instruct",
+  "architectures": [
+    "Idefics3ForConditionalGeneration"
+  ],
+  "image_seq_len": 81,
+  "image_token_id": 49153,
+  "model_type": "idefics3",
+  "pad_token_id": 128002,
+  "quantization_config": {
+    "config_groups": {
+      "group_0": {
+        "input_activations": null,
+        "output_activations": null,
+        "targets": [
+          "Linear"
+        ],
+        "weights": {
+          "actorder": null,
+          "block_structure": null,
+          "dynamic": false,
+          "group_size": 128,
+          "num_bits": 4,
+          "observer": "minmax",
+          "observer_kwargs": {},
+          "strategy": "group",
+          "symmetric": true,
+          "type": "int"
+        }
+      }
+    },
+    "format": "pack-quantized",
+    "global_compression_ratio": 1.3447756848086136,
+    "ignore": [
+      "model.vision_model.encoder.layers.0.self_attn.k_proj",
+      "model.vision_model.encoder.layers.0.self_attn.v_proj",
+      "model.vision_model.encoder.layers.0.self_attn.q_proj",
+      "model.vision_model.encoder.layers.0.self_attn.out_proj",
+      "model.vision_model.encoder.layers.0.mlp.fc1",
+      "model.vision_model.encoder.layers.0.mlp.fc2",
+      "model.vision_model.encoder.layers.1.self_attn.k_proj",
+      "model.vision_model.encoder.layers.1.self_attn.v_proj",
+      "model.vision_model.encoder.layers.1.self_attn.q_proj",
+      "model.vision_model.encoder.layers.1.self_attn.out_proj",
+      "model.vision_model.encoder.layers.1.mlp.fc1",
+      "model.vision_model.encoder.layers.1.mlp.fc2",
+      "model.vision_model.encoder.layers.2.self_attn.k_proj",
+      "model.vision_model.encoder.layers.2.self_attn.v_proj",
+      "model.vision_model.encoder.layers.2.self_attn.q_proj",
+      "model.vision_model.encoder.layers.2.self_attn.out_proj",
+      "model.vision_model.encoder.layers.2.mlp.fc1",
+      "model.vision_model.encoder.layers.2.mlp.fc2",
+      "model.vision_model.encoder.layers.3.self_attn.k_proj",
+      "model.vision_model.encoder.layers.3.self_attn.v_proj",
+      "model.vision_model.encoder.layers.3.self_attn.q_proj",
+      "model.vision_model.encoder.layers.3.self_attn.out_proj",
+      "model.vision_model.encoder.layers.3.mlp.fc1",
+      "model.vision_model.encoder.layers.3.mlp.fc2",
+      "model.vision_model.encoder.layers.4.self_attn.k_proj",
+      "model.vision_model.encoder.layers.4.self_attn.v_proj",
+      "model.vision_model.encoder.layers.4.self_attn.q_proj",
+      "model.vision_model.encoder.layers.4.self_attn.out_proj",
+      "model.vision_model.encoder.layers.4.mlp.fc1",
+      "model.vision_model.encoder.layers.4.mlp.fc2",
+      "model.vision_model.encoder.layers.5.self_attn.k_proj",
+      "model.vision_model.encoder.layers.5.self_attn.v_proj",
+      "model.vision_model.encoder.layers.5.self_attn.q_proj",
+      "model.vision_model.encoder.layers.5.self_attn.out_proj",
+      "model.vision_model.encoder.layers.5.mlp.fc1",
+      "model.vision_model.encoder.layers.5.mlp.fc2",
+      "model.vision_model.encoder.layers.6.self_attn.k_proj",
+      "model.vision_model.encoder.layers.6.self_attn.v_proj",
+      "model.vision_model.encoder.layers.6.self_attn.q_proj",
+      "model.vision_model.encoder.layers.6.self_attn.out_proj",
+      "model.vision_model.encoder.layers.6.mlp.fc1",
+      "model.vision_model.encoder.layers.6.mlp.fc2",
+      "model.vision_model.encoder.layers.7.self_attn.k_proj",
+      "model.vision_model.encoder.layers.7.self_attn.v_proj",
+      "model.vision_model.encoder.layers.7.self_attn.q_proj",
+      "model.vision_model.encoder.layers.7.self_attn.out_proj",
+      "model.vision_model.encoder.layers.7.mlp.fc1",
+      "model.vision_model.encoder.layers.7.mlp.fc2",
+      "model.vision_model.encoder.layers.8.self_attn.k_proj",
+      "model.vision_model.encoder.layers.8.self_attn.v_proj",
+      "model.vision_model.encoder.layers.8.self_attn.q_proj",
+      "model.vision_model.encoder.layers.8.self_attn.out_proj",
+      "model.vision_model.encoder.layers.8.mlp.fc1",
+      "model.vision_model.encoder.layers.8.mlp.fc2",
+      "model.vision_model.encoder.layers.9.self_attn.k_proj",
+      "model.vision_model.encoder.layers.9.self_attn.v_proj",
+      "model.vision_model.encoder.layers.9.self_attn.q_proj",
+      "model.vision_model.encoder.layers.9.self_attn.out_proj",
+      "model.vision_model.encoder.layers.9.mlp.fc1",
+      "model.vision_model.encoder.layers.9.mlp.fc2",
+      "model.vision_model.encoder.layers.10.self_attn.k_proj",
+      "model.vision_model.encoder.layers.10.self_attn.v_proj",
+      "model.vision_model.encoder.layers.10.self_attn.q_proj",
+      "model.vision_model.encoder.layers.10.self_attn.out_proj",
+      "model.vision_model.encoder.layers.10.mlp.fc1",
+      "model.vision_model.encoder.layers.10.mlp.fc2",
+      "model.vision_model.encoder.layers.11.self_attn.k_proj",
+      "model.vision_model.encoder.layers.11.self_attn.v_proj",
+      "model.vision_model.encoder.layers.11.self_attn.q_proj",
+      "model.vision_model.encoder.layers.11.self_attn.out_proj",
+      "model.vision_model.encoder.layers.11.mlp.fc1",
+      "model.vision_model.encoder.layers.11.mlp.fc2",
+      "model.vision_model.encoder.layers.12.self_attn.k_proj",
+      "model.vision_model.encoder.layers.12.self_attn.v_proj",
+      "model.vision_model.encoder.layers.12.self_attn.q_proj",
+      "model.vision_model.encoder.layers.12.self_attn.out_proj",
+      "model.vision_model.encoder.layers.12.mlp.fc1",
+      "model.vision_model.encoder.layers.12.mlp.fc2",
+      "model.vision_model.encoder.layers.13.self_attn.k_proj",
+      "model.vision_model.encoder.layers.13.self_attn.v_proj",
+      "model.vision_model.encoder.layers.13.self_attn.q_proj",
+      "model.vision_model.encoder.layers.13.self_attn.out_proj",
+      "model.vision_model.encoder.layers.13.mlp.fc1",
+      "model.vision_model.encoder.layers.13.mlp.fc2",
+      "model.vision_model.encoder.layers.14.self_attn.k_proj",
+      "model.vision_model.encoder.layers.14.self_attn.v_proj",
+      "model.vision_model.encoder.layers.14.self_attn.q_proj",
+      "model.vision_model.encoder.layers.14.self_attn.out_proj",
+      "model.vision_model.encoder.layers.14.mlp.fc1",
+      "model.vision_model.encoder.layers.14.mlp.fc2",
+      "model.vision_model.encoder.layers.15.self_attn.k_proj",
+      "model.vision_model.encoder.layers.15.self_attn.v_proj",
+      "model.vision_model.encoder.layers.15.self_attn.q_proj",
+      "model.vision_model.encoder.layers.15.self_attn.out_proj",
+      "model.vision_model.encoder.layers.15.mlp.fc1",
+      "model.vision_model.encoder.layers.15.mlp.fc2",
+      "model.vision_model.encoder.layers.16.self_attn.k_proj",
+      "model.vision_model.encoder.layers.16.self_attn.v_proj",
+      "model.vision_model.encoder.layers.16.self_attn.q_proj",
+      "model.vision_model.encoder.layers.16.self_attn.out_proj",
+      "model.vision_model.encoder.layers.16.mlp.fc1",
+      "model.vision_model.encoder.layers.16.mlp.fc2",
+      "model.vision_model.encoder.layers.17.self_attn.k_proj",
+      "model.vision_model.encoder.layers.17.self_attn.v_proj",
+      "model.vision_model.encoder.layers.17.self_attn.q_proj",
+      "model.vision_model.encoder.layers.17.self_attn.out_proj",
+      "model.vision_model.encoder.layers.17.mlp.fc1",
+      "model.vision_model.encoder.layers.17.mlp.fc2",
+      "model.vision_model.encoder.layers.18.self_attn.k_proj",
+      "model.vision_model.encoder.layers.18.self_attn.v_proj",
+      "model.vision_model.encoder.layers.18.self_attn.q_proj",
+      "model.vision_model.encoder.layers.18.self_attn.out_proj",
+      "model.vision_model.encoder.layers.18.mlp.fc1",
+      "model.vision_model.encoder.layers.18.mlp.fc2",
+      "model.vision_model.encoder.layers.19.self_attn.k_proj",
+      "model.vision_model.encoder.layers.19.self_attn.v_proj",
+      "model.vision_model.encoder.layers.19.self_attn.q_proj",
+      "model.vision_model.encoder.layers.19.self_attn.out_proj",
+      "model.vision_model.encoder.layers.19.mlp.fc1",
+      "model.vision_model.encoder.layers.19.mlp.fc2",
+      "model.vision_model.encoder.layers.20.self_attn.k_proj",
+      "model.vision_model.encoder.layers.20.self_attn.v_proj",
+      "model.vision_model.encoder.layers.20.self_attn.q_proj",
+      "model.vision_model.encoder.layers.20.self_attn.out_proj",
+      "model.vision_model.encoder.layers.20.mlp.fc1",
+      "model.vision_model.encoder.layers.20.mlp.fc2",
+      "model.vision_model.encoder.layers.21.self_attn.k_proj",
+      "model.vision_model.encoder.layers.21.self_attn.v_proj",
+      "model.vision_model.encoder.layers.21.self_attn.q_proj",
+      "model.vision_model.encoder.layers.21.self_attn.out_proj",
+      "model.vision_model.encoder.layers.21.mlp.fc1",
+      "model.vision_model.encoder.layers.21.mlp.fc2",
+      "model.vision_model.encoder.layers.22.self_attn.k_proj",
+      "model.vision_model.encoder.layers.22.self_attn.v_proj",
+      "model.vision_model.encoder.layers.22.self_attn.q_proj",
+      "model.vision_model.encoder.layers.22.self_attn.out_proj",
+      "model.vision_model.encoder.layers.22.mlp.fc1",
+      "model.vision_model.encoder.layers.22.mlp.fc2",
+      "model.vision_model.encoder.layers.23.self_attn.k_proj",
+      "model.vision_model.encoder.layers.23.self_attn.v_proj",
+      "model.vision_model.encoder.layers.23.self_attn.q_proj",
+      "model.vision_model.encoder.layers.23.self_attn.out_proj",
+      "model.vision_model.encoder.layers.23.mlp.fc1",
+      "model.vision_model.encoder.layers.23.mlp.fc2",
+      "model.vision_model.encoder.layers.24.self_attn.k_proj",
+      "model.vision_model.encoder.layers.24.self_attn.v_proj",
+      "model.vision_model.encoder.layers.24.self_attn.q_proj",
+      "model.vision_model.encoder.layers.24.self_attn.out_proj",
+      "model.vision_model.encoder.layers.24.mlp.fc1",
+      "model.vision_model.encoder.layers.24.mlp.fc2",
+      "model.vision_model.encoder.layers.25.self_attn.k_proj",
+      "model.vision_model.encoder.layers.25.self_attn.v_proj",
+      "model.vision_model.encoder.layers.25.self_attn.q_proj",
+      "model.vision_model.encoder.layers.25.self_attn.out_proj",
+      "model.vision_model.encoder.layers.25.mlp.fc1",
+      "model.vision_model.encoder.layers.25.mlp.fc2",
+      "model.vision_model.encoder.layers.26.self_attn.k_proj",
+      "model.vision_model.encoder.layers.26.self_attn.v_proj",
+      "model.vision_model.encoder.layers.26.self_attn.q_proj",
+      "model.vision_model.encoder.layers.26.self_attn.out_proj",
+      "model.vision_model.encoder.layers.26.mlp.fc1",
+      "model.vision_model.encoder.layers.26.mlp.fc2",
+      "model.connector.modality_projection.proj",
+      "lm_head"
+    ],
+    "kv_cache_scheme": null,
+    "quant_method": "compressed-tensors",
+    "quantization_status": "compressed",
+    "sparsity_config": {
+      "format": "dense",
+      "global_sparsity": 0.11567594517840002,
+      "ignore": [],
+      "registry_requires_subclass": false,
+      "sparsity_structure": "unstructured",
+      "targets": []
+    }
+  },
+  "scale_factor": 3,
+  "text_config": {
+    "_attn_implementation_autoset": true,
+    "_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",
+    "architectures": [
+      "VLlama3ForCausalLM"
+    ],
+    "bos_token_id": 0,
+    "eos_token_id": 0,
+    "head_dim": 64,
+    "hidden_size": 2048,
+    "intermediate_size": 8192,
+    "max_position_embeddings": 16384,
+    "model_type": "llama",
+    "neftune_noise_alpha": 0.0,
+    "num_hidden_layers": 24,
+    "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
+    },
+    "qk_layer_norms": false,
+    "rms_norm_eps": 1e-05,
+    "rope_theta": 273768.0,
+    "torch_dtype": "bfloat16",
+    "use_resampler": false,
+    "vocab_size": 49155
+  },
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers.js_config": {
+    "dtype": {
+      "decoder_model_merged": "q4",
+      "embed_tokens": "auto",
+      "vision_encoder": "auto"
+    },
+    "kv_cache_dtype": {
+      "fp16": "float16",
+      "q4f16": "float16"
+    }
+  },
+  "transformers_version": "4.48.1",
+  "use_cache": true,
+  "vision_config": {
+    "image_size": 384,
+    "intermediate_size": 4304,
+    "max_image_size": {
+      "longest_edge": 384
+    },
+    "model_type": "idefics3_vision",
+    "num_hidden_layers": 27,
+    "patch_size": 14,
+    "size": {
+      "longest_edge": 1920
+    },
+    "tie_word_embeddings": false
+  },
+  "vocab_size": 49155
+}

generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 0,
+  "eos_token_id": 49154,
+  "pad_token_id": 2,
+  "transformers_version": "4.48.1"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:830aa1f80eeca57b4d6c5d622e1befdf0153b06b68755467dabb430246844bb8
+size 2101921576

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"
+}

recipe.yaml

@@ -0,0 +1,7 @@
+DEFAULT_stage:
+  DEFAULT_modifiers:
+    GPTQModifier:
+      sequential_targets: [LlamaDecoderLayer]
+      scheme: W4A16
+      targets: Linear
+      ignore: ['re:.*lm_head', 're:model.vision_model.*', 're:model.connector.*']

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 +1,183 @@
+{
+  "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": "<|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 %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<end_of_utterance>",
+  "extra_special_tokens": {},
+  "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
+}