Upload folder using huggingface_hub

README.md

@@ -1,19 +1,23 @@
 ---
-license: apache-2.0
+base_model: Qwen/Qwen2-VL-2B-Instruct
 language:
 - en
+library_name: transformers
 pipeline_tag: image-text-to-text
+license: apache-2.0
 tags:
 - multimodal
+- qwen
+- qwen2
+- unsloth
+- transformers
+- vision
 - mlx
-library_name: transformers
-base_model:
-- Qwen/Qwen2-VL-2B
 ---
 
 # EZCon/Qwen2-VL-2B-Instruct-8bit-mlx
-This model was converted to MLX format from [`Qwen/Qwen2-VL-2B-Instruct`]() using mlx-vlm version **0.3.2**.
-Refer to the [original model card](https://huggingface.co/Qwen/Qwen2-VL-2B-Instruct) for more details on the model.
+This model was converted to MLX format from [`unsloth/Qwen2-VL-2B-Instruct`]() using mlx-vlm version **0.3.2**.
+Refer to the [original model card](https://huggingface.co/unsloth/Qwen2-VL-2B-Instruct) for more details on the model.
 ## Use with mlx
 
 ```bash

chat_template.json

@@ -1,3 +1,3 @@
 {
-    "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}"
-}
+  "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}"
+}

config.json

@@ -6,7 +6,7 @@
     "attention_dropout": 0.0,
     "bad_words_ids": null,
     "begin_suppress_tokens": null,
-    "bos_token_id": 151643,
+    "bos_token_id": null,
     "chunk_size_feed_forward": 0,
     "cross_attention_hidden_size": null,
     "decoder_start_token_id": null,
@@ -50,7 +50,7 @@
     "output_attentions": false,
     "output_hidden_states": false,
     "output_scores": false,
-    "pad_token_id": null,
+    "pad_token_id": 151654,
     "prefix": null,
     "problem_type": null,
     "pruned_heads": {},
@@ -1036,13 +1036,13 @@
     "return_dict_in_generate": false,
     "rms_norm_eps": 1e-06,
     "rope_scaling": {
-        "type": "default",
         "mrope_section": [
             16,
             24,
             24
         ],
-        "rope_type": "default"
+        "rope_type": "default",
+        "type": "default"
     },
     "rope_theta": 1000000.0,
     "sep_token_id": null,
@@ -1068,13 +1068,13 @@
         "rope_theta": 1000000.0,
         "attention_dropout": 0.0,
         "rope_scaling": {
-            "type": "default",
             "mrope_section": [
                 16,
                 24,
                 24
             ],
-            "rope_type": "default"
+            "rope_type": "default",
+            "type": "default"
         },
         "layer_types": [
             "full_attention",
@@ -1136,8 +1136,8 @@
         "problem_type": null,
         "tokenizer_class": null,
         "prefix": null,
-        "bos_token_id": 151643,
-        "pad_token_id": null,
+        "bos_token_id": null,
+        "pad_token_id": 151654,
         "eos_token_id": 151645,
         "sep_token_id": null,
         "decoder_start_token_id": null,
@@ -1166,11 +1166,12 @@
         "exponential_decay_length_penalty": null,
         "suppress_tokens": null,
         "begin_suppress_tokens": null,
-        "_name_or_path": "",
-        "vision_start_token_id": 151652,
+        "_name_or_path": "Qwen/Qwen2-VL-2B-Instruct",
+        "model_type": "qwen2_vl_text",
+        "unsloth_fixed": true,
         "vision_end_token_id": 151653,
+        "vision_start_token_id": 151652,
         "vision_token_id": 151654,
-        "model_type": "qwen2_vl_text",
         "tf_legacy_loss": false,
         "use_bfloat16": false,
         "output_attentions": false
@@ -1184,6 +1185,7 @@
     "torchscript": false,
     "transformers_version": "4.55.0.dev0",
     "typical_p": 1.0,
+    "unsloth_fixed": true,
     "use_bfloat16": false,
     "use_cache": true,
     "use_sliding_window": false,
@@ -1192,7 +1194,7 @@
         "return_dict": true,
         "output_hidden_states": false,
         "torchscript": false,
-        "torch_dtype": null,
+        "torch_dtype": "bfloat16",
         "pruned_heads": {},
         "tie_word_embeddings": true,
         "chunk_size_feed_forward": 0,
@@ -1247,6 +1249,7 @@
         "begin_suppress_tokens": null,
         "_name_or_path": "",
         "in_chans": 3,
+        "model_type": "qwen2_vl",
         "spatial_patch_size": 14,
         "tf_legacy_loss": false,
         "use_bfloat16": false,
@@ -1261,7 +1264,6 @@
         "spatial_merge_size": 2,
         "temporal_patch_size": 2,
         "initializer_range": 0.02,
-        "model_type": "qwen2_vl",
         "output_attentions": false
     },
     "vision_end_token_id": 151653,

generation_config.json

@@ -1,12 +1,14 @@
 {
-  "cache_implementation": "offloaded_hybrid_chunked",
-  "eos_token_id": 151645,
-  "forced_eos_token_id": 151645,
-  "low_memory": true,
-  "max_new_tokens": 8,
-  "min_new_tokens": 5,
-  "pad_token_id": 151643,
-  "remove_invalid_values": true,
-  "renormalize_logits": true,
-  "transformers_version": "4.55.0.dev0"
+  "bos_token_id": 151643,
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "max_length": 32768,
+  "pad_token_id": 151654,
+  "temperature": 0.01,
+  "top_k": 1,
+  "top_p": 0.001,
+  "transformers_version": "4.49.0"
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:21326527da58f616b6ca7e40b5866baee4e16c5375bc33473d86aa15130e922d
-size 2971011425
+oid sha256:641709b32bcc69c5c56c0ba31cc688ed118ecb626697dcbaa2f841135bc3db6b
+size 2971011801

special_tokens_map.json

@@ -22,7 +22,7 @@
     "single_word": false
   },
   "pad_token": {
-    "content": "<|endoftext|>",
+    "content": "<|vision_pad|>",
     "lstrip": false,
     "normalized": false,
     "rstrip": false,

tokenizer_config.json

@@ -135,7 +135,7 @@
   "errors": "replace",
   "extra_special_tokens": {},
   "model_max_length": 32768,
-  "pad_token": "<|endoftext|>",
+  "pad_token": "<|vision_pad|>",
   "padding_side": "left",
   "processor_class": "Qwen2VLProcessor",
   "split_special_tokens": false,