Upload folder using huggingface_hub

config.json

@@ -0,0 +1,69 @@
+{
+  "architectures": [
+    "OpenCUAForConditionalGeneration"
+  ],
+  "auto_map": {
+    "AutoConfig": "configuration_opencua.OpenCUAConfig",
+    "AutoModel": "modeling_opencua.OpenCUAForConditionalGeneration",
+    "AutoModelForCausalLM": "modeling_opencua.OpenCUAForConditionalGeneration"
+  },
+  "ignore_index": -100,
+  "media_placeholder_token_id": 151664,
+  "model_type": "opencua",
+  "pad_token_id": 0,
+  "text_config": {
+    "bos_token_id": 151643,
+    "eos_token_id": 151644,
+    "head_dim": 128,
+    "hidden_act": "silu",
+    "hidden_size": 8192,
+    "initializer_range": 0.02,
+    "intermediate_size": 29568,
+    "k_proj_bias": true,
+    "max_length": 20,
+    "min_length": 0,
+    "model_type": "qwen2",
+    "num_attention_heads": 64,
+    "num_beam_groups": 1,
+    "num_beams": 1,
+    "num_hidden_layers": 80,
+    "num_key_value_heads": 8,
+    "pad_token_id": 152063,
+    "pretraining_sequence_length": 131072,
+    "q_proj_bias": true,
+    "rms_norm_eps": 1e-05,
+    "rope_theta": 1000000.0,
+    "tie_word_embeddings": false,
+    "torch_dtype": "bfloat16",
+    "use_bfloat16": false,
+    "use_cache": true,
+    "v_proj_bias": true,
+    "vocab_size": 152064
+  },
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.48.3",
+  "vision_config": {
+    "depth": 32,
+    "fullatt_block_indexes": [
+      7,
+      15,
+      23,
+      31
+    ],
+    "hidden_act": "silu",
+    "hidden_size": 1280,
+    "num_heads": 16,
+    "in_chans": 3,
+    "intermediate_size": 3456,
+    
+    "patch_size": 14,
+    "spatial_merge_size": 2,
+    "spatial_patch_size": 14,
+    "temporal_patch_size": 2,
+    "out_hidden_size": 8192,
+    "tokens_per_second": 2,
+    "window_size": 112
+  },
+  "vocab_size": 152064
+}

configuration_opencua.py

@@ -0,0 +1,37 @@
+from transformers.configuration_utils import PretrainedConfig
+from transformers.models.qwen2_5_vl.configuration_qwen2_5_vl import Qwen2_5_VLVisionConfig
+from transformers.models.qwen2.configuration_qwen2 import Qwen2Config
+
+
+class OpenCUAConfig(PretrainedConfig):
+    """OpenCUA-2.5-32B model configuration.
+
+    Args:
+        vision_config: Configuration for the vision model.Qwen2_5_VLVisionConfig
+        text_config: Configuration for the text model. Qwen2Config
+        pad_token_id: The token ID to use for padding.
+    """
+
+    model_type = "opencua"
+
+    def __init__(
+        self,
+        vision_config: dict | Qwen2_5_VLVisionConfig | None = None,
+        text_config: dict | Qwen2Config | None = None,
+        ignore_index: int = -100,
+        media_placeholder_token_id: int = 151664,
+        pad_token_id: int = 0,
+        **kwargs
+    ):
+        if isinstance(vision_config, dict):
+            vision_config = Qwen2_5_VLVisionConfig(**vision_config)
+        self.vision_config = vision_config
+
+        if isinstance(text_config, dict):
+            text_config = Qwen2Config(**text_config)
+        self.text_config = text_config
+
+        self.ignore_index = ignore_index
+        self.media_placeholder_token_id = media_placeholder_token_id
+
+        super().__init__(pad_token_id=pad_token_id, **kwargs)

generation_config.json

@@ -0,0 +1,4 @@
+{
+  "max_length": 32768,
+  "eos_token_id": 151644
+}

model-1-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2d056b52a8a473d7a67bf2fd998d7d995be3a97df569d4750854c1a268194b2
+size 1755371048

model-10-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1d31c113efa0c422a90825344e62686721032c681ce9167a37c62035c3ec25c8
+size 1755371048

model-11-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2a8935673de0f1f3d25ffcdf0b237640a07797c7e3a6072710a5f89ed3d2008e
+size 1755371056

model-12-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6964c24544d368740553b347c822ddbde7b0d3dd754e8a20d2d6c828d86192d7
+size 1755371056

model-13-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:08a26184b23f8d30516082d4a95f9de51328220a871c91ffbdb35c7b02d5c3a5
+size 1755371056

model-14-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a342b259b2a9ac9b6664e77bdcde8846f096cf8f8a448ec92f862d01fc3a7d5e
+size 1755371056

model-15-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6b6faa37fdf65daa4f5ccd587c1885e3fd93d48062db8a224b40930339d55d2
+size 1755371056

model-16-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:89b9e97834fb2c820ec6c69e2149c4b28b245d4a9b38fd613471cb8c303db01e
+size 1755371056

model-17-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e121184a6eb22b40ce9d0667effb87ef8008a3eab679160b7052b2126dbdaa83
+size 1755371056

model-18-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:21f8de6571c5f2b208e02b0ced8a9ef136851ca7b3e973ea3928311c806ec3af
+size 1755371056

model-19-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:88651735a288765556bd3de077e3f6ea8dd7cb3ba462f09d32d92382b16576e3
+size 1755371056

model-2-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6013f340bd5ab57b778b395655a4abec0a5f682d7cb37d6f18af68a7b35db451
+size 1755371048

model-20-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c4131d77aac96479eb50471c9094c0820725419a969baa8f9c441de6b8e54187
+size 1755371056

model-21-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3d5dcf22b91c558273dbd4de9b7c21214156d0b910021eb72360ed887e0c3d9d
+size 1755371056

model-22-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:244f6eafe07220b99bb5ff8cfa1cb57535baea26cc4bd665633161e5643e396b
+size 1755371056

model-23-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1c641241e076557a74cce7f15987f50e9774b54b9b272014dcfceb42e7dcd57b
+size 1755371056

model-24-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:279747122fdd60c99f814ff378fd58f9e2e9fd124b9f32554aa7c50ca4c03c2c
+size 1755371056

model-25-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4f3af84c17e8c776817fd08309f855557e2c6628613063397355fca854ff3721
+size 1755371056

model-26-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:31c561ec3f44390e3833a06767e6fc5f36b3d5495e5a827dea6bdf39151a20bc
+size 1755371056

model-27-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be77ff47fe6e62b75f4d20cce69306ab140c9517c686a1389de705d5a5cb7250
+size 1755371056

model-28-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57193974fd8685b40ae607d26476ac69c5fe478c5b66a74ab9ac3afffef436a4
+size 1755371056

model-29-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:337a8f4208cd9470b1ad092e0ac64a757efa29c6970bf3ab3afe577616429ef9
+size 1755371056

model-3-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:387ed3f58de3cb274bf3d02e4392d07bf1782fde1c0796c990dc106a4902aefc
+size 1755371048

model-30-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:195a2c4636e46f1beaf5ab713f3edefbd63f103b19f5d51dbcfc9e7c9d8c4f78
+size 1755371056

model-31-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5b4a687312bd119f474148ba93a8aeadc9de6cac84755998fb55cd349085fd0a
+size 1755371056

model-32-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dea3e9a2365b3624462b26c031ba7c656e701be744ee077d0f210f0f88fd0ac5
+size 1755371056

model-33-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:81c57f93db82a0ddef33caae1a1af86328f8e9bafedc98df828eea121c017ca3
+size 1755371056

model-34-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ff4e8a0736363b9d3933162f5d5307478912795bb687d4e77a69dfac0e36687c
+size 1755371056

model-35-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:315126c76bb299fdd1c1689cd7a070c0469e01d5bcc44da90ec476e901b5e028
+size 1755371056

model-36-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e888c45e248799fe15d64981b111bb91b046355a39fbac8955f2319d4d815c5e
+size 1755371056

model-37-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f5eb3c97600f01de04250b502189abb350b5a46d4563b2fd768f691abfed2e3
+size 1755371056

model-38-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a75f96e8dc26d0436f30c2c3585fb430e84a841888adcc6b38b44ea8a8a26372
+size 1755371056

model-39-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8360f1df14d5e34d27cee56cf4cd8eaa0132737c33d7c8ecf0523ea0f7d2fbd3
+size 1755371056

model-4-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae1103b205f4174372af1df1ec16c8b50e4e2a5f56a6abe057bb67f3fce9f515
+size 1755371048

model-40-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a60533d810230f9c87c02b568efbc9cc1bd143cec80bcdb0bdaa9c4ee07a9e55
+size 1755371056

model-41-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fa377da8e7fc678236306f258f9c3e72472c5934e2998aad7811130186008495
+size 1755371056

model-42-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e161be31b8fef8d5e757ce54ea2b5c457efd42e5d0a22364d2e4039f983b886
+size 1755371056

model-43-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f1ed935c0d317084a88eb9cdd53cf0bd0655e1b355931e14e89b703f1a6c0ca
+size 1755371056

model-44-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1558912235eccf0055484dbe989cfd96db6b9bf3d12b638d78d44c2452f8bc35
+size 1755371056

model-45-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a94b605ed59966adae7663ffd2fb8cbb89ddf76cf42c8e2bbc70d1e4305a330
+size 1755371056

model-46-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e7af26dd6c148f7eebb7b56064a938c32ccb359c4c5e433150eb47fd2175e949
+size 1755371056

model-47-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c7772cc6a1efcb654b0375f7241edc6a76577dff64e0d0f31f6885cc7cb26dca
+size 1755371056

model-48-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5e9cc970204965198157f1dfc0ebaf5a8cab312e026527e52e718d9ab493e087
+size 1755371056

model-49-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d1d3f912f57824537f53b0e3fa137798dc6f82eedb0eac42bc2806446dd7841
+size 1755371056

model-5-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:132ef8bc3a773adcd987fc843c4a25834562752aae69ac4a9ab2ae84cef8b04e
+size 1755371048

model-50-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a12cf3789cfd8bf49e9d46be117fed5f32a8915fea01b63d88f800b98c81446
+size 1755371056

model-51-of-82.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01ca5afad959953aa24f5e06569cbb97e1b42e85d966a8cae9a141cc62eb34f4
+size 1755371056