Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,28 @@
+---
+license: apache-2.0
+library_name: mlx
+tags:
+- dllm
+- diffusion
+- llm
+- text_generation
+- mlx
+pipeline_tag: image-text-to-text
+base_model: inclusionAI/LLaDA2.1-mini
+---
+
+# mlx-community/LLaDA2.1-mini-bf16
+
+This model was converted to MLX format from [`inclusionAI/LLaDA2.1-mini`](https://huggingface.co/inclusionAI/LLaDA2.1-mini)
+using mlx-vlm version **0.5.0**.
+Refer to the [original model card](https://huggingface.co/inclusionAI/LLaDA2.1-mini) for more details on the model.
+
+## Use with mlx
+
+```bash
+pip install -U mlx-vlm
+```
+
+```bash
+python -m mlx_vlm.generate --model mlx-community/LLaDA2.1-mini-bf16 --max-tokens 100 --temperature 0.0 --prompt "Describe this image." --image <path_to_image>
+```

chat_template.jinja

@@ -0,0 +1,86 @@
+{% set thinking_option = 'off' %}
+{{- '<role>SYSTEM</role>' }}
+{%- if messages[0].role == 'system' %}
+    {{- messages[0].content + '\n' }}
+{%- endif %}
+{%- if tools %}
+    {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call>\n" }}
+{%- endif %}
+{{- 'detailed thinking ' + thinking_option + '<|role_end|>' }}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if message.content is string %}
+        {%- set content = message.content %}
+    {%- else %}
+        {%- set content = '' %}
+    {%- endif %}
+    {%- if message.role == "user" %}
+        {{- '<role>HUMAN</role>' + message.content + '<|role_end|>' }}
+    {%- elif message.role == "system" and not loop.first %}
+        {{- '<role>SYSTEM</role>' + message.content + '<|role_end|>' }}
+    {%- elif message.role == "assistant" %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is string %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in content %}
+                {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+                {%- set content = content.split('</think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if reasoning_content %}
+                {{- '<role>ASSISTANT</role>' + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<role>ASSISTANT</role>' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<role>ASSISTANT</role>' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|role_end|>' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<role>OBSERVATION</role>' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|role_end|>' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<role>ASSISTANT</role>' }}
+{%- endif %}

config.json

@@ -0,0 +1,56 @@
+{
+    "architectures": [
+        "LLaDA2MoeModelLM"
+    ],
+    "attention_dropout": 0.0,
+    "auto_map": {
+        "AutoConfig": "configuration_llada2_moe.LLaDA2MoeConfig",
+        "AutoModel": "modeling_llada2_moe.LLaDA2MoeModel",
+        "AutoModelForCausalLM": "modeling_llada2_moe.LLaDA2MoeModelLM"
+    },
+    "dtype": "bfloat16",
+    "embedding_dropout": 0.0,
+    "first_k_dense_replace": 1,
+    "head_dim": 128,
+    "hidden_act": "silu",
+    "hidden_size": 2048,
+    "initializer_range": 0.02,
+    "intermediate_size": 5120,
+    "max_position_embeddings": 32768,
+    "max_window_layers": 28,
+    "model_type": "llada2_moe",
+    "moe_intermediate_size": 512,
+    "moe_router_enable_expert_bias": true,
+    "n_group": 8,
+    "norm_head": false,
+    "norm_softmax": false,
+    "norm_topk_prob": true,
+    "num_attention_heads": 16,
+    "num_experts": 256,
+    "num_experts_per_tok": 8,
+    "num_hidden_layers": 20,
+    "num_key_value_heads": 4,
+    "num_shared_experts": 1,
+    "output_dropout": 0.0,
+    "output_router_logits": false,
+    "pad_token_id": 156892,
+    "partial_rotary_factor": 0.5,
+    "rms_norm_eps": 1e-06,
+    "rope_scaling": null,
+    "rope_theta": 600000,
+    "rotary_dim": 64,
+    "routed_scaling_factor": 2.5,
+    "router_dtype": "fp32",
+    "score_function": "sigmoid",
+    "sliding_window": 4096,
+    "tie_word_embeddings": false,
+    "topk_group": 4,
+    "transformers_version": "4.57.1",
+    "use_bias": false,
+    "use_cache": false,
+    "use_qkv_bias": false,
+    "use_rmsnorm": true,
+    "use_sliding_window": false,
+    "using_split_qkv_in_self_attention": false,
+    "vocab_size": 157184
+}

configuration_llada2_moe.py

@@ -0,0 +1,88 @@
+"""LLaDA2 MoE model configuration"""
+
+from transformers.configuration_utils import PretrainedConfig
+
+
+class LLaDA2MoeConfig(PretrainedConfig):
+    model_type = "llada2_moe"
+
+    def __init__(
+        self,
+        vocab_size=30592,
+        hidden_size=1024,
+        intermediate_size=None,
+        num_hidden_layers=24,
+        num_attention_heads=16,
+        num_key_value_heads=0,
+        hidden_act="silu",
+        use_qkv_bias=False,  # llada2 only
+        use_qk_norm=True,
+        use_bias=True,  # llada2 only
+        rms_norm_eps=1e-05,
+        norm_head=False,  # llada2 only
+        tie_word_embeddings=False,  # PretrainedConfig key, here change default value.
+        embedding_dropout=0.1,
+        attention_dropout=0.1,
+        output_dropout=0.1,
+        initializer_range=0.02,
+        max_position_embeddings=16384,
+        rope_theta=10000.0,
+        use_cache=True,
+        use_sliding_window=False,
+        sliding_window=4096,
+        max_window_layers=28,
+        rope_scaling=None,
+        pad_token_id=126081,
+        num_experts=16,
+        num_shared_experts=0,
+        num_experts_per_tok=2,
+        n_group=8,
+        topk_group=4,
+        routed_scaling_factor=2.5,
+        moe_intermediate_size=None,
+        first_k_dense_replace=0,
+        head_dim=None,
+        output_router_logits=False,
+        partial_rotary_factor=0.5,
+        **kwargs,
+    ):
+        self.num_hidden_layers = num_hidden_layers
+        self.vocab_size = vocab_size
+        self.hidden_size = hidden_size
+        self.intermediate_size = intermediate_size
+        self.num_attention_heads = num_attention_heads
+        self.num_key_value_heads = num_key_value_heads
+        self.hidden_act = hidden_act
+        self.use_qkv_bias = use_qkv_bias
+        self.use_qk_norm = use_qk_norm
+        self.use_bias = use_bias
+        self.norm_head = norm_head
+        self.rms_norm_eps = rms_norm_eps
+        self.embedding_dropout = embedding_dropout
+        self.attention_dropout = attention_dropout
+        self.output_dropout = output_dropout
+        self.initializer_range = initializer_range
+        self.max_position_embeddings = max_position_embeddings
+        self.rope_theta = rope_theta
+        self.use_cache = use_cache
+        self.use_sliding_window = use_sliding_window
+        self.sliding_window = sliding_window
+        self.max_window_layers = max_window_layers
+        self.head_dim = head_dim or self.hidden_size // self.num_attention_heads
+        self.rope_scaling = rope_scaling
+
+        # MoE configs
+        self.num_experts = num_experts
+        self.num_shared_experts = num_shared_experts
+        self.num_experts_per_tok = num_experts_per_tok
+        self.n_group = n_group
+        self.topk_group = topk_group
+        self.moe_intermediate_size = moe_intermediate_size
+        self.first_k_dense_replace = first_k_dense_replace
+        self.output_router_logits = output_router_logits
+        self.routed_scaling_factor = routed_scaling_factor
+        self.partial_rotary_factor = partial_rotary_factor
+
+        super().__init__(
+            pad_token_id=pad_token_id, tie_word_embeddings=tie_word_embeddings, **kwargs
+        )

model-00001-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d7edffdd3250525755c83f3d077b5a40dcaa334cdb1b3a97e1917f950ac36af4
+size 5101358123

model-00002-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:69066d94bd59c59e20c6097506606b0115e384d85f3563822f2e838048581c07
+size 4916807732

model-00003-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:647566728b497ff6814f0aeaf02942c88a94ecdb1909f2331fe90b6abe3baf80
+size 4916807764

model-00004-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a5602a45b544d784cc648059f9309b1e04ec6685094bed583f9c1dcb31161aa6
+size 4916807792

model-00005-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5cfafd2faf20ce945b0c6a344077a331625fd21bba13579b74023f45b3d059e8
+size 4916807778

model-00006-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:460da7281094f14779a82d40cbe8782aecb311fcd5be091359a0d6af71d4b46f
+size 4916807778

model-00007-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ae4e3575d13a2457fdeff7a44fab04ba6789c8b97cba3f1fab121c000bd375e
+size 2825937321

model.safetensors.index.json

@@ -0,0 +1,285 @@
+{
+    "metadata": {
+        "total_size": 32511296512
+    },
+    "weight_map": {
+        "language_model.lm_head.weight": "model-00007-of-00007.safetensors",
+        "language_model.model.layers.0.attention.dense.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.0.attention.key_layernorm.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.0.attention.query_key_value.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.0.attention.query_layernorm.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.0.input_layernorm.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.0.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.0.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.0.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.0.post_attention_layernorm.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.1.attention.dense.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.1.attention.key_layernorm.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.1.attention.query_key_value.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.1.attention.query_layernorm.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.1.input_layernorm.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.1.mlp.gate.expert_bias": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.1.mlp.gate.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.1.mlp.shared_experts.down_proj.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.1.mlp.shared_experts.gate_proj.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.1.mlp.shared_experts.up_proj.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.1.mlp.switch_mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.1.mlp.switch_mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.1.mlp.switch_mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.1.post_attention_layernorm.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.10.attention.dense.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.10.attention.key_layernorm.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.10.attention.query_key_value.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.10.attention.query_layernorm.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.10.input_layernorm.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.10.mlp.gate.expert_bias": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.10.mlp.gate.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.10.mlp.shared_experts.down_proj.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.10.mlp.shared_experts.gate_proj.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.10.mlp.shared_experts.up_proj.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.10.mlp.switch_mlp.down_proj.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.10.mlp.switch_mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.10.mlp.switch_mlp.up_proj.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.10.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.11.attention.dense.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.11.attention.key_layernorm.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.11.attention.query_key_value.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.11.attention.query_layernorm.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.11.input_layernorm.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.11.mlp.gate.expert_bias": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.11.mlp.gate.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.11.mlp.shared_experts.down_proj.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.11.mlp.shared_experts.gate_proj.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.11.mlp.shared_experts.up_proj.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.11.mlp.switch_mlp.down_proj.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.11.mlp.switch_mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.11.mlp.switch_mlp.up_proj.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.11.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.12.attention.dense.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.12.attention.key_layernorm.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.12.attention.query_key_value.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.12.attention.query_layernorm.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.12.input_layernorm.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.12.mlp.gate.expert_bias": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.12.mlp.gate.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.12.mlp.shared_experts.down_proj.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.12.mlp.shared_experts.gate_proj.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.12.mlp.shared_experts.up_proj.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.12.mlp.switch_mlp.down_proj.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.12.mlp.switch_mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.12.mlp.switch_mlp.up_proj.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.12.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.13.attention.dense.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.13.attention.key_layernorm.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.13.attention.query_key_value.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.13.attention.query_layernorm.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.13.input_layernorm.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.13.mlp.gate.expert_bias": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.13.mlp.gate.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.13.mlp.shared_experts.down_proj.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.13.mlp.shared_experts.gate_proj.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.13.mlp.shared_experts.up_proj.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.13.mlp.switch_mlp.down_proj.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.13.mlp.switch_mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.13.mlp.switch_mlp.up_proj.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.13.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.14.attention.dense.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.14.attention.key_layernorm.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.14.attention.query_key_value.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.14.attention.query_layernorm.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.14.input_layernorm.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.14.mlp.gate.expert_bias": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.14.mlp.gate.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.14.mlp.shared_experts.down_proj.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.14.mlp.shared_experts.gate_proj.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.14.mlp.shared_experts.up_proj.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.14.mlp.switch_mlp.down_proj.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.14.mlp.switch_mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.14.mlp.switch_mlp.up_proj.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.14.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.15.attention.dense.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.15.attention.key_layernorm.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.15.attention.query_key_value.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.15.attention.query_layernorm.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.15.input_layernorm.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.15.mlp.gate.expert_bias": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.15.mlp.gate.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.15.mlp.shared_experts.down_proj.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.15.mlp.shared_experts.gate_proj.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.15.mlp.shared_experts.up_proj.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.15.mlp.switch_mlp.down_proj.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.15.mlp.switch_mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.15.mlp.switch_mlp.up_proj.weight": "model-00005-of-00007.safetensors",
+        "language_model.model.layers.15.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.16.attention.dense.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.16.attention.key_layernorm.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.16.attention.query_key_value.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.16.attention.query_layernorm.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.16.input_layernorm.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.16.mlp.gate.expert_bias": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.16.mlp.gate.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.16.mlp.shared_experts.down_proj.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.16.mlp.shared_experts.gate_proj.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.16.mlp.shared_experts.up_proj.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.16.mlp.switch_mlp.down_proj.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.16.mlp.switch_mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.16.mlp.switch_mlp.up_proj.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.16.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.17.attention.dense.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.17.attention.key_layernorm.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.17.attention.query_key_value.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.17.attention.query_layernorm.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.17.input_layernorm.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.17.mlp.gate.expert_bias": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.17.mlp.gate.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.17.mlp.shared_experts.down_proj.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.17.mlp.shared_experts.gate_proj.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.17.mlp.shared_experts.up_proj.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.17.mlp.switch_mlp.down_proj.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.17.mlp.switch_mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.17.mlp.switch_mlp.up_proj.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.17.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.18.attention.dense.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.18.attention.key_layernorm.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.18.attention.query_key_value.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.18.attention.query_layernorm.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.18.input_layernorm.weight": "model-00007-of-00007.safetensors",
+        "language_model.model.layers.18.mlp.gate.expert_bias": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.18.mlp.gate.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.18.mlp.shared_experts.down_proj.weight": "model-00007-of-00007.safetensors",
+        "language_model.model.layers.18.mlp.shared_experts.gate_proj.weight": "model-00007-of-00007.safetensors",
+        "language_model.model.layers.18.mlp.shared_experts.up_proj.weight": "model-00007-of-00007.safetensors",
+        "language_model.model.layers.18.mlp.switch_mlp.down_proj.weight": "model-00007-of-00007.safetensors",
+        "language_model.model.layers.18.mlp.switch_mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.18.mlp.switch_mlp.up_proj.weight": "model-00006-of-00007.safetensors",
+        "language_model.model.layers.18.post_attention_layernorm.weight": "model-00007-of-00007.safetensors",
+        "language_model.model.layers.19.attention.dense.weight": "model-00007-of-00007.safetensors",
+        "language_model.model.layers.19.attention.key_layernorm.weight": "model-00007-of-00007.safetensors",
+        "language_model.model.layers.19.attention.query_key_value.weight": "model-00007-of-00007.safetensors",
+        "language_model.model.layers.19.attention.query_layernorm.weight": "model-00007-of-00007.safetensors",
+        "language_model.model.layers.19.input_layernorm.weight": "model-00007-of-00007.safetensors",
+        "language_model.model.layers.19.mlp.gate.expert_bias": "model-00007-of-00007.safetensors",
+        "language_model.model.layers.19.mlp.gate.weight": "model-00007-of-00007.safetensors",
+        "language_model.model.layers.19.mlp.shared_experts.down_proj.weight": "model-00007-of-00007.safetensors",
+        "language_model.model.layers.19.mlp.shared_experts.gate_proj.weight": "model-00007-of-00007.safetensors",
+        "language_model.model.layers.19.mlp.shared_experts.up_proj.weight": "model-00007-of-00007.safetensors",
+        "language_model.model.layers.19.mlp.switch_mlp.down_proj.weight": "model-00007-of-00007.safetensors",
+        "language_model.model.layers.19.mlp.switch_mlp.gate_proj.weight": "model-00007-of-00007.safetensors",
+        "language_model.model.layers.19.mlp.switch_mlp.up_proj.weight": "model-00007-of-00007.safetensors",
+        "language_model.model.layers.19.post_attention_layernorm.weight": "model-00007-of-00007.safetensors",
+        "language_model.model.layers.2.attention.dense.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.2.attention.key_layernorm.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.2.attention.query_key_value.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.2.attention.query_layernorm.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.2.input_layernorm.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.2.mlp.gate.expert_bias": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.2.mlp.gate.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.2.mlp.shared_experts.down_proj.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.2.mlp.shared_experts.gate_proj.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.2.mlp.shared_experts.up_proj.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.2.mlp.switch_mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.2.mlp.switch_mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.2.mlp.switch_mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.2.post_attention_layernorm.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.3.attention.dense.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.3.attention.key_layernorm.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.3.attention.query_key_value.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.3.attention.query_layernorm.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.3.input_layernorm.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.3.mlp.gate.expert_bias": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.3.mlp.gate.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.3.mlp.shared_experts.down_proj.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.3.mlp.shared_experts.gate_proj.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.3.mlp.shared_experts.up_proj.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.3.mlp.switch_mlp.down_proj.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.3.mlp.switch_mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.3.mlp.switch_mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+        "language_model.model.layers.3.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.4.attention.dense.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.4.attention.key_layernorm.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.4.attention.query_key_value.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.4.attention.query_layernorm.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.4.input_layernorm.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.4.mlp.gate.expert_bias": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.4.mlp.gate.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.4.mlp.shared_experts.down_proj.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.4.mlp.shared_experts.gate_proj.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.4.mlp.shared_experts.up_proj.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.4.mlp.switch_mlp.down_proj.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.4.mlp.switch_mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.4.mlp.switch_mlp.up_proj.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.4.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.5.attention.dense.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.5.attention.key_layernorm.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.5.attention.query_key_value.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.5.attention.query_layernorm.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.5.input_layernorm.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.5.mlp.gate.expert_bias": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.5.mlp.gate.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.5.mlp.shared_experts.down_proj.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.5.mlp.shared_experts.gate_proj.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.5.mlp.shared_experts.up_proj.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.5.mlp.switch_mlp.down_proj.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.5.mlp.switch_mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.5.mlp.switch_mlp.up_proj.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.5.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.6.attention.dense.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.6.attention.key_layernorm.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.6.attention.query_key_value.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.6.attention.query_layernorm.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.6.input_layernorm.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.6.mlp.gate.expert_bias": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.6.mlp.gate.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.6.mlp.shared_experts.down_proj.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.6.mlp.shared_experts.gate_proj.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.6.mlp.shared_experts.up_proj.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.6.mlp.switch_mlp.down_proj.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.6.mlp.switch_mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.6.mlp.switch_mlp.up_proj.weight": "model-00002-of-00007.safetensors",
+        "language_model.model.layers.6.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.7.attention.dense.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.7.attention.key_layernorm.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.7.attention.query_key_value.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.7.attention.query_layernorm.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.7.input_layernorm.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.7.mlp.gate.expert_bias": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.7.mlp.gate.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.7.mlp.shared_experts.down_proj.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.7.mlp.shared_experts.gate_proj.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.7.mlp.shared_experts.up_proj.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.7.mlp.switch_mlp.down_proj.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.7.mlp.switch_mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.7.mlp.switch_mlp.up_proj.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.7.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.8.attention.dense.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.8.attention.key_layernorm.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.8.attention.query_key_value.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.8.attention.query_layernorm.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.8.input_layernorm.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.8.mlp.gate.expert_bias": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.8.mlp.gate.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.8.mlp.shared_experts.down_proj.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.8.mlp.shared_experts.gate_proj.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.8.mlp.shared_experts.up_proj.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.8.mlp.switch_mlp.down_proj.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.8.mlp.switch_mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.8.mlp.switch_mlp.up_proj.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.8.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.9.attention.dense.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.9.attention.key_layernorm.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.9.attention.query_key_value.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.9.attention.query_layernorm.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.9.input_layernorm.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.9.mlp.gate.expert_bias": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.9.mlp.gate.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.9.mlp.shared_experts.down_proj.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.9.mlp.shared_experts.gate_proj.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.9.mlp.shared_experts.up_proj.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.9.mlp.switch_mlp.down_proj.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.layers.9.mlp.switch_mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.9.mlp.switch_mlp.up_proj.weight": "model-00003-of-00007.safetensors",
+        "language_model.model.layers.9.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
+        "language_model.model.norm.weight": "model-00007-of-00007.safetensors",
+        "language_model.model.word_embeddings.weight": "model-00001-of-00007.safetensors"
+    }
+}

modeling_llada2_moe.py

@@ -0,0 +1,1439 @@
+# Copyright 2025 Antgroup and The HuggingFace Inc. team. All rights reserved.
+#
+# This code is based on EleutherAI's GPT-NeoX library and the GPT-NeoX
+# and OPT implementations in this library. It has been modified from its
+# original forms to accommodate minor architectural differences compared
+# to GPT-NeoX and OPT used by the Meta AI team that trained the model.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""PyTorch LLaDA2MoE model."""
+
+import math
+from typing import List, Callable, Optional, Tuple, Union
+
+import torch
+import torch.nn.functional as F
+from torch import nn
+from torch.nn import CrossEntropyLoss
+
+from transformers.activations import ACT2FN
+from transformers.cache_utils import Cache, DynamicCache
+from transformers.masking_utils import create_bidirectional_mask
+from transformers.modeling_outputs import (
+    MoeModelOutputWithPast,
+    MoeCausalLMOutputWithPast,
+)
+from transformers.modeling_rope_utils import ROPE_INIT_FUNCTIONS, dynamic_rope_update
+from transformers.modeling_utils import ALL_ATTENTION_FUNCTIONS, PreTrainedModel
+from transformers.processing_utils import Unpack
+from transformers.pytorch_utils import (
+    ALL_LAYERNORM_LAYERS,
+)
+from transformers.utils import (
+    TransformersKwargs,
+    add_start_docstrings,
+    add_start_docstrings_to_model_forward,
+    logging,
+    replace_return_docstrings,
+)
+from .configuration_llada2_moe import LLaDA2MoeConfig
+from transformers.generation.utils import GenerationMixin
+
+
+logger = logging.get_logger(__name__)
+
+_CONFIG_FOR_DOC = "LLaDA2MoeConfig"
+
+
+def _get_unpad_data(attention_mask):
+    seqlens_in_batch = attention_mask.sum(dim=-1, dtype=torch.int32)
+    indices = torch.nonzero(attention_mask.flatten(), as_tuple=False).flatten()
+    max_seqlen_in_batch = seqlens_in_batch.max().item()
+    cu_seqlens = F.pad(
+        torch.cumsum(seqlens_in_batch, dim=0, dtype=torch.torch.int32), (1, 0)
+    )
+    return (
+        indices,
+        cu_seqlens,
+        max_seqlen_in_batch,
+    )
+
+
+class LLaDA2MoeRMSNorm(nn.Module):
+    def __init__(self, hidden_size, eps=1e-6):
+        """
+        LLaDA2MoeRMSNorm is equivalent to T5LayerNorm
+        """
+        super().__init__()
+        self.weight = nn.Parameter(torch.ones(hidden_size))
+        self.variance_epsilon = eps
+
+    def forward(self, hidden_states):
+        input_dtype = hidden_states.dtype
+        hidden_states = hidden_states.to(torch.float32)
+        variance = hidden_states.pow(2).mean(-1, keepdim=True)
+        hidden_states = hidden_states * torch.rsqrt(variance + self.variance_epsilon)
+        return self.weight * hidden_states.to(input_dtype)
+
+
+ALL_LAYERNORM_LAYERS.append(LLaDA2MoeRMSNorm)
+
+
+class LLaDA2MoeRotaryEmbedding(nn.Module):
+    inv_freq: torch.Tensor  # fix linting for register_buffer
+
+    def __init__(self, config: LLaDA2MoeConfig, device=None):
+        super().__init__()
+        self.max_seq_len_cached = config.max_position_embeddings
+        self.original_max_seq_len = config.max_position_embeddings
+
+        self.config = config
+
+        self.rope_type = self.config.rope_parameters["rope_type"]
+        rope_init_fn: Callable = self.compute_default_rope_parameters
+        if self.rope_type != "default":
+            rope_init_fn = ROPE_INIT_FUNCTIONS[self.rope_type]
+        inv_freq, self.attention_scaling = rope_init_fn(self.config, device)
+
+        self.register_buffer("inv_freq", inv_freq, persistent=False)
+        self.register_buffer("original_inv_freq", inv_freq.clone(), persistent=False)
+
+    @staticmethod
+    def compute_default_rope_parameters(
+        config: LLaDA2MoeConfig = None,
+        device=None,
+        seq_len: int = None,
+    ):
+        base = config.rope_parameters["rope_theta"]
+        partial_rotary_factor = config.rope_parameters.get("partial_rotary_factor", 1.0)
+        head_dim = getattr(config, "head_dim", None) or config.hidden_size // config.num_attention_heads
+        dim = int(head_dim * partial_rotary_factor)
+
+        attention_factor = 1.0  # Unused in this type of RoPE
+
+        inv_freq = 1.0 / (
+            base ** (torch.arange(0, dim, 2, dtype=torch.int64).to(device=device, dtype=torch.float) / dim)
+        )
+        return inv_freq, attention_factor
+
+    @torch.no_grad()
+    @dynamic_rope_update  # power user: used with advanced RoPE types (e.g. dynamic rope)
+    def forward(self, x, position_ids):
+        inv_freq_expanded = (
+            self.inv_freq[None, :, None]
+            .float()
+            .expand(position_ids.shape[0], -1, 1)
+            .to(x.device)
+        )
+        position_ids_expanded = position_ids[:, None, :].float()
+
+        device_type = (
+            x.device.type
+            if isinstance(x.device.type, str) and x.device.type != "mps"
+            else "cpu"
+        )
+        with torch.autocast(device_type=device_type, enabled=False):  # Force float32
+            freqs = (
+                inv_freq_expanded.float() @ position_ids_expanded.float()
+            ).transpose(1, 2)
+            emb = torch.cat((freqs, freqs), dim=-1)
+            cos = emb.cos() * self.attention_scaling
+            sin = emb.sin() * self.attention_scaling
+
+        return cos.to(dtype=x.dtype), sin.to(dtype=x.dtype)
+
+
+# Copied from transformers.models.llama.modeling_llama.rotate_half
+def rotate_half(x):
+    """Rotates half the hidden dims of the input."""
+    x1 = x[..., : x.shape[-1] // 2]
+    x2 = x[..., x.shape[-1] // 2 :]
+    return torch.cat((-x2, x1), dim=-1)
+
+
+# Copied from transformers.models.llama.modeling_llama.apply_rotary_pos_emb
+def apply_rotary_pos_emb(q, k, cos, sin, position_ids=None, unsqueeze_dim=1):
+    """Applies Rotary Position Embedding to the query and key tensors.
+
+    Args:
+        q (`torch.Tensor`): The query tensor.
+        k (`torch.Tensor`): The key tensor.
+        cos (`torch.Tensor`): The cosine part of the rotary embedding.
+        sin (`torch.Tensor`): The sine part of the rotary embedding.
+        position_ids (`torch.Tensor`):
+            The position indices of the tokens corresponding to the query and key tensors. For example, this can be
+            used to pass offsetted position ids when working with a KV-cache.
+        unsqueeze_dim (`int`, *optional*, defaults to 1):
+            The 'unsqueeze_dim' argument specifies the dimension along which to unsqueeze cos[position_ids] and
+            sin[position_ids] so that they can be properly broadcasted to the dimensions of q and k. For example, note
+            that cos[position_ids] and sin[position_ids] have the shape [batch_size, seq_len, head_dim]. Then, if q and
+            k have the shape [batch_size, heads, seq_len, head_dim], then setting unsqueeze_dim=1 makes
+            cos[position_ids] and sin[position_ids] broadcastable to the shapes of q and k. Similarly, if q and k have
+            the shape [batch_size, seq_len, heads, head_dim], then set unsqueeze_dim=2.
+    Returns:
+        `tuple(torch.Tensor)` comprising the query and key tensors rotated using the Rotary Position Embedding.
+    """
+    cos = cos.unsqueeze(unsqueeze_dim)
+    sin = sin.unsqueeze(unsqueeze_dim)
+
+    # Keep half or full tensor for later concatenation
+    rotary_dim = cos.shape[-1]
+    q_rot, q_pass = q[..., :rotary_dim], q[..., rotary_dim:]
+    k_rot, k_pass = k[..., :rotary_dim], k[..., rotary_dim:]
+
+    # Apply rotary embeddings on the first half or full tensor
+    q_embed = (q_rot * cos) + (rotate_half(q_rot) * sin)
+    k_embed = (k_rot * cos) + (rotate_half(k_rot) * sin)
+
+    # Concatenate back to full shape
+    q_embed = torch.cat([q_embed, q_pass], dim=-1)
+    k_embed = torch.cat([k_embed, k_pass], dim=-1)
+    return q_embed, k_embed
+
+
+class LLaDA2MoeMLP(nn.Module):
+    def __init__(self, config: LLaDA2MoeConfig, intermediate_size: int):
+        super().__init__()
+        self.config = config
+        self.hidden_size = config.hidden_size
+        self.intermediate_size = intermediate_size
+
+        self.gate_proj = nn.Linear(self.hidden_size, self.intermediate_size, bias=False)
+        self.up_proj = nn.Linear(self.hidden_size, self.intermediate_size, bias=False)
+        self.down_proj = nn.Linear(self.intermediate_size, self.hidden_size, bias=False)
+        self.act_fn = ACT2FN[config.hidden_act]
+
+    def forward(self, x):
+        return self.down_proj(self.act_fn(self.gate_proj(x)) * self.up_proj(x))
+
+
+class LLaDA2MoeGate(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.config = config
+        self.top_k = config.num_experts_per_tok
+        self.num_experts = config.num_experts
+
+        self.n_group = config.n_group
+        self.topk_group = config.topk_group
+
+        # topk selection algorithm
+        self.gating_dim = config.hidden_size
+        self.weight = nn.Parameter(torch.empty((self.num_experts, self.gating_dim)))
+        self.routed_scaling_factor = config.routed_scaling_factor
+
+        self.register_buffer("expert_bias", torch.zeros(self.num_experts))
+        self.reset_parameters()
+
+    def reset_parameters(self) -> None:
+        import torch.nn.init as init
+
+        init.kaiming_uniform_(self.weight, a=math.sqrt(5))
+
+    def group_limited_topk(
+        self,
+        scores: torch.Tensor,
+    ):
+        num_tokens, _ = scores.size()
+        # Organize the experts into groups
+        group_scores = (
+            scores.view(num_tokens, self.n_group, -1).topk(2, dim=-1)[0].sum(dim=-1)
+        )
+        group_idx = torch.topk(group_scores, k=self.topk_group, dim=-1, sorted=False)[1]
+        group_mask = torch.zeros_like(group_scores)
+        group_mask.scatter_(1, group_idx, 1)
+
+        # Mask the experts based on selection groups
+        score_mask = (
+            group_mask.unsqueeze(-1)
+            .expand(num_tokens, self.n_group, self.num_experts // self.n_group)
+            .reshape(num_tokens, -1)
+        )
+
+        masked_scores = scores.masked_fill(~score_mask.bool(), float("-inf"))
+        probs, top_indices = torch.topk(masked_scores, k=self.top_k, dim=-1)
+
+        return probs, top_indices
+
+    def forward(self, hidden_states):
+        # compute gating score
+        hidden_states = hidden_states.view(-1, hidden_states.shape[-1])
+        logits = F.linear(
+            hidden_states.type(torch.float32), self.weight.type(torch.float32)
+        )
+
+        scores = torch.sigmoid(logits.float()).type_as(logits)
+
+        scores_for_routing = scores + self.expert_bias
+        _, topk_idx = self.group_limited_topk(scores_for_routing)
+
+        scores = torch.gather(scores, dim=1, index=topk_idx).type_as(logits)
+
+        topk_weight = (
+            scores / (scores.sum(dim=-1, keepdim=True) + 1e-20)
+            if self.top_k > 1
+            else scores
+        )
+        topk_weight = topk_weight * self.routed_scaling_factor
+
+        return topk_idx, topk_weight, logits
+
+
+class LLaDA2MoeSparseMoeBlock(nn.Module):
+    """
+    A mixed expert module containing shared experts.
+    """
+
+    def __init__(self, config: LLaDA2MoeConfig):
+        super().__init__()
+        self.config = config
+        self.num_experts_per_tok = config.num_experts_per_tok
+        self._setup_experts()
+        self.gate = LLaDA2MoeGate(config)
+        if config.num_shared_experts is not None:
+            self.shared_experts = LLaDA2MoeMLP(
+                config=config,
+                intermediate_size=config.moe_intermediate_size
+                * config.num_shared_experts,
+            )
+
+    def _setup_experts(self):
+        self.experts = nn.ModuleList(
+            [
+                LLaDA2MoeMLP(
+                    config=self.config,
+                    intermediate_size=self.config.moe_intermediate_size,
+                )
+                for _ in range(self.config.num_experts)
+            ]
+        )
+
+    def forward(self, hidden_states):
+        identity = hidden_states
+        bsz, seq_len, h = hidden_states.shape
+        topk_idx, topk_weight, router_logits = self.gate(hidden_states)
+        hidden_states = hidden_states.view(-1, hidden_states.shape[-1])
+        flat_topk_idx = topk_idx.view(-1)
+        if self.training:
+            hidden_states = hidden_states.repeat_interleave(
+                self.num_experts_per_tok, dim=0
+            )
+            y = torch.empty_like(hidden_states)
+            for i, expert in enumerate(self.experts):
+                y[flat_topk_idx == i] = expert(hidden_states[flat_topk_idx == i])
+            y = (y.view(*topk_weight.shape, -1) * topk_weight.unsqueeze(-1)).sum(dim=1)
+            y = y.to(hidden_states.dtype).view(bsz, seq_len, h)
+        else:
+            y = self.moe_infer(hidden_states, topk_idx, topk_weight).view(
+                bsz, seq_len, h
+            )
+        if self.config.num_shared_experts is not None:
+            y = y + self.shared_experts(identity)
+        return y, (
+            router_logits.view(bsz, seq_len, -1),
+            topk_idx.view(bsz, seq_len, -1),
+        )
+
+    @torch.no_grad()
+    def moe_infer(self, x, topk_ids, topk_weight):
+        cnts = topk_ids.new_zeros((topk_ids.shape[0], len(self.experts)))
+        cnts.scatter_(1, topk_ids, 1)
+        tokens_per_expert = cnts.sum(dim=0)
+        idxs = topk_ids.view(-1).argsort()
+        sorted_tokens = x[idxs // topk_ids.shape[1]]
+        tokens_per_expert = tokens_per_expert.cpu().numpy()
+        outputs = []
+        start_idx = 0
+        for i, num_tokens_tensor in enumerate(tokens_per_expert):
+            num_tokens = num_tokens_tensor.item()
+            if num_tokens == 0:
+                continue
+            end_idx = start_idx + num_tokens
+            expert = self.experts[i]
+            tokens_for_this_expert = sorted_tokens[start_idx:end_idx]
+            expert_out = expert(tokens_for_this_expert)
+            outputs.append(expert_out.to(x.device))
+            start_idx = end_idx
+
+        outs = torch.cat(outputs, dim=0) if len(outputs) else sorted_tokens.new_empty(0)
+        new_x = torch.empty_like(outs)
+        new_x[idxs] = outs
+        final_out = (
+            new_x.view(*topk_ids.shape, -1)
+            .type(topk_weight.dtype)
+            .mul_(topk_weight.unsqueeze(dim=-1))
+            .sum(dim=1)
+            .type(new_x.dtype)
+        )
+        return final_out
+
+
+# Copied from transformers.models.llama.modeling_llama.repeat_kv
+def repeat_kv(hidden_states: torch.Tensor, n_rep: int) -> torch.Tensor:
+    """
+    This is the equivalent of torch.repeat_interleave(x, dim=1, repeats=n_rep). The hidden states go from (batch,
+    num_key_value_heads, seqlen, head_dim) to (batch, num_attention_heads, seqlen, head_dim)
+    """
+    batch, num_key_value_heads, slen, head_dim = hidden_states.shape
+    if n_rep == 1:
+        return hidden_states
+    hidden_states = hidden_states[:, :, None, :, :].expand(
+        batch, num_key_value_heads, n_rep, slen, head_dim
+    )
+    return hidden_states.reshape(batch, num_key_value_heads * n_rep, slen, head_dim)
+
+
+def eager_attention_forward(
+    module: nn.Module,
+    query: torch.Tensor,
+    key: torch.Tensor,
+    value: torch.Tensor,
+    attention_mask: Optional[torch.Tensor],
+    scaling: float,
+    dropout: float = 0.0,
+    **kwargs: Unpack[TransformersKwargs],
+):
+    key_states = repeat_kv(key, module.num_key_value_groups)
+    value_states = repeat_kv(value, module.num_key_value_groups)
+
+    attn_weights = torch.matmul(query, key_states.transpose(2, 3)) * scaling
+    if attention_mask is not None:
+        attn_weights = attn_weights + attention_mask[:, :, :, : key_states.shape[-2]]
+
+    # upcast attention to fp32
+    attn_weights = nn.functional.softmax(attn_weights, dim=-1, dtype=torch.float32).to(
+        query.dtype
+    )
+    attn_weights = nn.functional.dropout(
+        attn_weights, p=dropout, training=module.training
+    )
+    attn_output = torch.matmul(attn_weights, value_states)
+    attn_output = attn_output.transpose(1, 2).contiguous()
+
+    return attn_output, attn_weights
+
+
+# Copied from transformers.models.llama.modeling_llama.LlamaAttention with Llama->LLaDA2Moe
+class LLaDA2MoeAttention(nn.Module):
+    """Multi-headed attention from 'Attention Is All You Need' paper"""
+
+    def __init__(self, config: LLaDA2MoeConfig, layer_idx: Optional[int] = None):
+        super().__init__()
+        self.config = config
+        self.layer_idx = layer_idx
+        if layer_idx is None:
+            logger.warning_once(
+                f"Instantiating {self.__class__.__name__} without passing `layer_idx` is not recommended and will "
+                "to errors during the forward call, if caching is used. Please make sure to provide a `layer_idx` "
+                "when creating this class."
+            )
+        self.attention_dropout = config.attention_dropout
+        self.hidden_size = config.hidden_size
+        self.num_heads = config.num_attention_heads
+        self.head_dim = config.head_dim or self.hidden_size // self.num_heads
+        partial_rotary_factor = (
+            config.partial_rotary_factor
+            if hasattr(config, "partial_rotary_factor")
+            else 1.0
+        )
+        self.rope_dim = int(self.head_dim * partial_rotary_factor)
+        self.num_key_value_heads = config.num_key_value_heads
+        self.num_key_value_groups = self.num_heads // self.num_key_value_heads
+        self.max_position_embeddings = config.max_position_embeddings
+        self.rope_theta = config.rope_theta
+        self.scaling = self.head_dim**-0.5
+        self.is_causal = False
+
+        self.query_key_value = nn.Linear(
+            self.hidden_size,
+            (self.num_heads + 2 * self.num_key_value_heads) * self.head_dim,
+            bias=config.use_qkv_bias,
+        )
+
+        if self.config.use_qk_norm:
+            self.query_layernorm = LLaDA2MoeRMSNorm(
+                self.head_dim, eps=config.rms_norm_eps
+            )
+            self.key_layernorm = LLaDA2MoeRMSNorm(
+                self.head_dim, eps=config.rms_norm_eps
+            )
+        self.dense = nn.Linear(
+            self.num_heads * self.head_dim, self.hidden_size, bias=config.use_bias
+        )
+        self.sliding_window = getattr(config, "sliding_window", None)
+
+    def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
+        return (
+            tensor.view(bsz, seq_len, self.num_heads, self.head_dim)
+            .transpose(1, 2)
+            .contiguous()
+        )
+
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_value: Optional[Cache] = None,
+        output_attentions: bool = False,
+        use_cache: bool = False,
+        position_embeddings: Optional[
+            Tuple[torch.Tensor, torch.Tensor]
+        ] = None,  # necessary, but kept here for BC
+        **kwargs,
+    ) -> Tuple[torch.Tensor, Optional[torch.Tensor], Optional[Tuple[torch.Tensor]]]:
+        input_shape = hidden_states.shape[:-1]
+
+        bsz, q_len, _ = hidden_states.size()
+
+        qkv = self.query_key_value(hidden_states)
+        qkv = qkv.view(
+            bsz, q_len, self.num_heads + 2 * self.num_key_value_heads, self.head_dim
+        )
+
+        query_states, key_states, value_states = qkv.split(
+            [self.num_heads, self.num_key_value_heads, self.num_key_value_heads], dim=-2
+        )
+        query_states = query_states.transpose(1, 2)
+        key_states = key_states.transpose(1, 2)
+        value_states = value_states.transpose(1, 2)
+
+        if self.config.use_qk_norm:
+            query_states = self.query_layernorm(query_states)
+            key_states = self.key_layernorm(key_states)
+
+        cos, sin = position_embeddings
+        query_states, key_states = apply_rotary_pos_emb(
+            query_states, key_states, cos, sin
+        )
+
+        if past_key_value is not None:
+            if self.layer_idx is None:
+                raise ValueError(
+                    f"The cache structure has changed since version v4.36. If you are using {self.__class__.__name__} "
+                    "for auto-regressive decoding with k/v caching, please make sure to initialize the attention class "
+                    "with a layer index."
+                )
+            cache_kwargs = {"sin": sin, "cos": cos}
+            key_states, value_states = past_key_value.update(
+                key_states, value_states, self.layer_idx, cache_kwargs
+            )
+
+        attention_interface: Callable = eager_attention_forward
+        if self.config._attn_implementation != "eager":
+            attention_interface = ALL_ATTENTION_FUNCTIONS[
+                self.config._attn_implementation
+            ]
+
+        attn_output, attn_weights = attention_interface(
+            self,
+            query_states,
+            key_states,
+            value_states,
+            attention_mask,
+            dropout=0.0 if not self.training else self.attention_dropout,
+            scaling=self.scaling,
+            sliding_window=self.sliding_window,  # diff with Llama
+            **kwargs,
+        )
+
+        attn_output = attn_output.reshape(*input_shape, -1).contiguous()
+        attn_output = self.dense(attn_output)
+
+        return attn_output, attn_weights, past_key_value
+
+
+class LLaDA2MoeDecoderLayer(nn.Module):
+    def __init__(self, config: LLaDA2MoeConfig, layer_idx: int):
+        super().__init__()
+        self.hidden_size = config.hidden_size
+
+        self.attention = LLaDA2MoeAttention(config=config, layer_idx=layer_idx)
+
+        self.mlp = (
+            LLaDA2MoeSparseMoeBlock(config)
+            if (
+                config.num_experts is not None
+                and layer_idx >= config.first_k_dense_replace
+            )
+            else LLaDA2MoeMLP(config=config, intermediate_size=config.intermediate_size)
+        )
+        self.input_layernorm = LLaDA2MoeRMSNorm(
+            config.hidden_size, eps=config.rms_norm_eps
+        )
+        self.post_attention_layernorm = LLaDA2MoeRMSNorm(
+            config.hidden_size, eps=config.rms_norm_eps
+        )
+
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_value: Optional[Tuple[torch.Tensor]] = None,
+        output_attentions: Optional[bool] = False,
+        output_router_logits: Optional[bool] = False,
+        use_cache: Optional[bool] = False,
+        position_embeddings: Optional[
+            Tuple[torch.Tensor, torch.Tensor]
+        ] = None,  # necessary, but kept here for BC
+        **kwargs,
+    ) -> Tuple[
+        torch.FloatTensor, Optional[Tuple[torch.FloatTensor, torch.FloatTensor]]
+    ]:
+        """
+        Args:
+            hidden_states (`torch.FloatTensor`): input to the layer of shape `(batch, seq_len, embed_dim)`
+            attention_mask (`torch.FloatTensor`, *optional*):
+                attention mask of size `(batch_size, sequence_length)` if flash attention is used or `(batch_size, 1,
+                query_sequence_length, key_sequence_length)` if default attention is used.
+            position_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
+                Indices of positions of each input sequence tokens in the position embeddings. Selected in the range `[0,
+                config.n_positions - 1]`.
+            past_key_value (`Tuple(torch.FloatTensor)`, *optional*):
+                cached past key and value projection states
+            output_attentions (`bool`, *optional*):
+                Whether to return the attentions tensors of all attention layers. See `attentions` under
+                returned tensors for more detail.
+            output_router_logits (`bool`, *optional*):
+                Whether or not to return the logits of all the routers. They are useful for computing the router loss,
+                and should not be returned during inference.
+            use_cache (`bool`, *optional*):
+                If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding
+                (see `past_key_values`).
+        """
+        residual = hidden_states
+
+        hidden_states = self.input_layernorm(hidden_states)
+
+        # Self Attention
+        hidden_states, self_attn_weights, present_key_value = self.attention(
+            hidden_states=hidden_states,
+            attention_mask=attention_mask,
+            position_ids=position_ids,
+            past_key_value=past_key_value,
+            output_attentions=output_attentions,
+            position_embeddings=position_embeddings,
+            use_cache=use_cache,
+        )
+        hidden_states = residual + hidden_states
+
+        # Fully Connected
+        residual = hidden_states
+        hidden_states = self.post_attention_layernorm(hidden_states)
+        hidden_states = self.mlp(hidden_states)
+        if isinstance(hidden_states, tuple):
+            hidden_states, router_logits = hidden_states
+        else:
+            router_logits = None
+        hidden_states = residual + hidden_states.to(residual.device)
+
+        outputs = (hidden_states,)
+
+        if output_attentions:
+            outputs += (self_attn_weights,)
+
+        if use_cache:
+            outputs += (present_key_value,)
+
+        if output_router_logits:
+            outputs += (router_logits,)
+
+        return outputs
+
+
+LLADA2MOE_START_DOCSTRING = r"""
+    This model inherits from [`PreTrainedModel`]. Check the superclass documentation for the generic methods the
+    library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads
+    etc.)
+
+    This model is also a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) subclass.
+    Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage
+    and behavior.
+
+    Parameters:
+        config ([`LLaDA2MoeConfig`]):
+            Model configuration class with all the parameters of the model. Initializing with a config file does not
+            load the weights associated with the model, only the configuration. Check out the
+            [`~PreTrainedModel.from_pretrained`] method to load the model weights.
+"""
+
+
+@add_start_docstrings(
+    "The bare LLaDA2Moe Model outputting raw hidden-states without any specific head on top.",
+    LLADA2MOE_START_DOCSTRING,
+)
+class LLaDA2MoePreTrainedModel(PreTrainedModel):
+    config_class = LLaDA2MoeConfig
+    base_model_prefix = "model"
+    supports_gradient_checkpointing = True
+    _no_split_modules = ["LLaDA2MoeDecoderLayer"]
+    _skip_keys_device_placement = ["past_key_values"]
+    _supports_flash_attn_2 = False
+    _supports_sdpa = True
+    _supports_flex_attn = True
+    _supports_cache_class = True
+
+    @torch.no_grad()
+    def _init_weights(self, module):
+        super()._init_weights(module)
+        std = self.config.initializer_range
+        if isinstance(module, LLaDA2MoeGate):
+            nn.init.normal_(module.weight, mean=0.0, std=std)
+
+
+LLADA2MOE_INPUTS_DOCSTRING = r"""
+    Args:
+        input_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`):
+            Indices of input sequence tokens in the vocabulary. Padding will be ignored by default should you provide
+            it.
+
+            Indices can be obtained using [`AutoTokenizer`]. See [`PreTrainedTokenizer.encode`] and
+            [`PreTrainedTokenizer.__call__`] for details.
+
+            [What are input IDs?](../glossary#input-ids)
+        attention_mask (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*):
+            Mask to avoid performing attention on padding token indices. Mask values selected in `[0, 1]`:
+
+            - 1 for tokens that are **not masked**,
+            - 0 for tokens that are **masked**.
+
+            [What are attention masks?](../glossary#attention-mask)
+
+            Indices can be obtained using [`AutoTokenizer`]. See [`PreTrainedTokenizer.encode`] and
+            [`PreTrainedTokenizer.__call__`] for details.
+
+            If `past_key_values` is used, optionally only the last `input_ids` have to be input (see
+            `past_key_values`).
+
+            If you want to change padding behavior, you should read [`modeling_opt._prepare_decoder_attention_mask`]
+            and modify to your needs. See diagram 1 in [the paper](https://arxiv.org/abs/1910.13461) for more
+            information on the default strategy.
+
+            - 1 indicates the head is **not masked**,
+            - 0 indicates the head is **masked**.
+        position_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
+            Indices of positions of each input sequence tokens in the position embeddings. Selected in the range `[0,
+            config.n_positions - 1]`.
+
+            [What are position IDs?](../glossary#position-ids)
+        past_key_values (`Cache` or `tuple(tuple(torch.FloatTensor))`, *optional*):
+            Pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention
+            blocks) that can be used to speed up sequential decoding. This typically consists in the `past_key_values`
+            returned by the model at a previous stage of decoding, when `use_cache=True` or `config.use_cache=True`.
+
+            Two formats are allowed:
+            - a [`~cache_utils.Cache`] instance;
+            - Tuple of `tuple(torch.FloatTensor)` of length `config.n_layers`, with each tuple having 2 tensors of
+            shape `(batch_size, num_heads, sequence_length, embed_size_per_head)`). This is also known as the legacy
+            cache format.
+
+            The model will output the same cache format that is fed as input. If no `past_key_values` are passed, the
+            legacy cache format will be returned.
+
+            If `past_key_values` are used, the user can optionally input only the last `input_ids` (those that don't
+            have their past key value states given to this model) of shape `(batch_size, 1)` instead of all `input_ids`
+            of shape `(batch_size, sequence_length)`.
+        inputs_embeds (`torch.FloatTensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*):
+            Optionally, instead of passing `input_ids` you can choose to directly pass an embedded representation. This
+            is useful if you want more control over how to convert `input_ids` indices into associated vectors than the
+            model's internal embedding lookup matrix.
+        use_cache (`bool`, *optional*):
+            If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding (see
+            `past_key_values`).
+        output_attentions (`bool`, *optional*):
+            Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned
+            tensors for more detail.
+        output_hidden_states (`bool`, *optional*):
+            Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for
+            more detail.
+        return_dict (`bool`, *optional*):
+            Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
+"""
+
+
+@add_start_docstrings(
+    "The bare LLaDA2Moe Model outputting raw hidden-states without any specific head on top.",
+    LLADA2MOE_START_DOCSTRING,
+)
+class LLaDA2MoeModel(LLaDA2MoePreTrainedModel):
+    """
+    Transformer decoder consisting of *config.num_hidden_layers* layers. Each layer is a [`LLaDA2MoeDecoderLayer`]
+
+    Args:
+        config: LLaDA2MoeConfig
+    """
+
+    def __init__(self, config: LLaDA2MoeConfig):
+        super().__init__(config)
+        self.padding_idx = config.pad_token_id
+        self.vocab_size = config.vocab_size
+
+        self.word_embeddings = nn.Embedding(
+            config.vocab_size, config.hidden_size, self.padding_idx
+        )
+        self.layers = nn.ModuleList(
+            [
+                LLaDA2MoeDecoderLayer(config, layer_idx)
+                for layer_idx in range(config.num_hidden_layers)
+            ]
+        )
+        self._use_sdpa = config._attn_implementation == "sdpa"
+        self._use_flex_attention = config._attn_implementation == "flex_attention"
+        self.norm = LLaDA2MoeRMSNorm(config.hidden_size, eps=config.rms_norm_eps)
+        self.rotary_emb = LLaDA2MoeRotaryEmbedding(config=config)
+        self.gradient_checkpointing = False
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    def get_input_embeddings(self):
+        return self.word_embeddings
+
+    def set_input_embeddings(self, value):
+        self.word_embeddings = value
+
+    @add_start_docstrings_to_model_forward(LLADA2MOE_INPUTS_DOCSTRING)
+    def forward(
+        self,
+        input_ids: torch.LongTensor = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_values: Optional[List[torch.FloatTensor]] = None,
+        inputs_embeds: Optional[torch.FloatTensor] = None,
+        use_cache: Optional[bool] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        output_router_logits: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+        **kwargs,
+    ) -> Union[Tuple, MoeModelOutputWithPast]:
+        output_attentions = (
+            output_attentions
+            if output_attentions is not None
+            else self.config.output_attentions
+        )
+        output_hidden_states = (
+            output_hidden_states
+            if output_hidden_states is not None
+            else self.config.output_hidden_states
+        )
+        output_router_logits = (
+            output_router_logits
+            if output_router_logits is not None
+            else self.config.output_router_logits
+        )
+        use_cache = use_cache if use_cache is not None else self.config.use_cache
+
+        return_dict = (
+            return_dict if return_dict is not None else self.config.use_return_dict
+        )
+
+        # retrieve input_ids and inputs_embeds
+        if input_ids is not None and inputs_embeds is not None:
+            raise ValueError(
+                "You cannot specify both input_ids and inputs_embeds at the same time"
+            )
+        elif input_ids is not None:
+            batch_size, seq_length = input_ids.shape[:2]
+        elif inputs_embeds is not None:
+            batch_size, seq_length = inputs_embeds.shape[:2]
+        else:
+            raise ValueError("You have to specify either input_ids or inputs_embeds")
+
+        if self.gradient_checkpointing and self.training:
+            if use_cache:
+                logger.warning_once(
+                    "`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`transformers."
+                )
+                use_cache = False
+
+        if use_cache and past_key_values is None:
+            past_key_values = DynamicCache()
+
+        if inputs_embeds is None:
+            inputs_embeds = self.word_embeddings(input_ids)
+
+        past_seen_tokens = (
+            past_key_values.get_seq_length() if past_key_values is not None else 0
+        )
+
+        if position_ids is None:
+            position_ids = torch.arange(
+                past_seen_tokens,
+                past_seen_tokens + inputs_embeds.shape[1],
+                device=inputs_embeds.device,
+            )
+            position_ids = position_ids.unsqueeze(0)
+        attention_mask = create_bidirectional_mask(
+            config=self.config,
+            inputs_embeds=inputs_embeds,
+            attention_mask=attention_mask,
+        )
+        # embed positions
+        hidden_states = inputs_embeds
+
+        # create position embeddings to be shared across the decoder layers
+        position_embeddings = self.rotary_emb(hidden_states, position_ids)
+
+        # decoder layers
+        all_hidden_states = () if output_hidden_states else None
+        all_self_attns = () if output_attentions else None
+        all_router_logits = () if output_router_logits else None
+        next_decoder_cache = None
+
+        for decoder_layer in self.layers:
+            if output_hidden_states:
+                all_hidden_states += (hidden_states,)
+
+            if self.gradient_checkpointing and self.training:
+                layer_outputs = self._gradient_checkpointing_func(
+                    decoder_layer.__call__,
+                    hidden_states,
+                    attention_mask,
+                    position_ids,
+                    past_key_values,
+                    output_attentions,
+                    output_router_logits,
+                    use_cache,
+                    position_embeddings,
+                )
+            else:
+                layer_outputs = decoder_layer(
+                    hidden_states,
+                    attention_mask=attention_mask,
+                    position_ids=position_ids,
+                    past_key_value=past_key_values,
+                    output_attentions=output_attentions,
+                    output_router_logits=output_router_logits,
+                    use_cache=use_cache,
+                    position_embeddings=position_embeddings,
+                )
+            hidden_states = layer_outputs[0]
+
+            if use_cache:
+                next_decoder_cache = layer_outputs[2 if output_attentions else 1]
+
+            if output_attentions:
+                all_self_attns += (layer_outputs[1],)
+
+            if output_router_logits and layer_outputs[-1] is not None:
+                all_router_logits += (layer_outputs[-1],)
+
+        hidden_states = self.norm(hidden_states)
+
+        # add hidden states from the last decoder layer
+        if output_hidden_states:
+            all_hidden_states += (hidden_states,)
+
+        next_cache = None
+        if use_cache:
+            next_cache = next_decoder_cache
+        if not return_dict:
+            return tuple(
+                v
+                for v in [
+                    hidden_states,
+                    next_cache,
+                    all_hidden_states,
+                    all_self_attns,
+                    all_router_logits,
+                ]
+                if v is not None
+            )
+        return MoeModelOutputWithPast(
+            last_hidden_state=hidden_states,
+            past_key_values=next_cache,
+            hidden_states=all_hidden_states,
+            attentions=all_self_attns,
+            router_logits=all_router_logits,
+        )
+
+
+class LLaDA2MoeModelLM(LLaDA2MoePreTrainedModel, GenerationMixin):
+    _tied_weights_keys = ["lm_head.weight"]
+
+    def __init__(self, config: LLaDA2MoeConfig):
+        super().__init__(config)
+        self.model = LLaDA2MoeModel(config)
+        self.vocab_size = config.vocab_size
+        self.lm_head = nn.Linear(config.hidden_size, config.vocab_size, bias=False)
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    def get_input_embeddings(self):
+        return self.model.word_embeddings
+
+    def set_input_embeddings(self, value):
+        self.model.word_embeddings = value
+
+    def get_output_embeddings(self):
+        return self.lm_head
+
+    def set_output_embeddings(self, new_embeddings):
+        self.lm_head = new_embeddings
+
+    def set_decoder(self, decoder):
+        self.model = decoder
+
+    def get_decoder(self):
+        return self.model
+
+    @add_start_docstrings_to_model_forward(LLADA2MOE_INPUTS_DOCSTRING)
+    @replace_return_docstrings(
+        output_type=MoeCausalLMOutputWithPast, config_class=_CONFIG_FOR_DOC
+    )
+    def forward(
+        self,
+        input_ids: torch.LongTensor = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_values: Optional[List[torch.FloatTensor]] = None,
+        inputs_embeds: Optional[torch.FloatTensor] = None,
+        labels: Optional[torch.LongTensor] = None,
+        use_cache: Optional[bool] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        output_router_logits: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+        **kwargs,
+    ) -> Union[Tuple, MoeCausalLMOutputWithPast]:
+        r"""
+        Args:
+            labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
+                Labels for computing the masked language modeling loss. Indices should either be in `[0, ...,
+                config.vocab_size]` or -100 (see `input_ids` docstring). Tokens with indices set to `-100` are ignored
+                (masked), the loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`.
+
+        Returns:
+
+        Example:
+
+        ```python
+        >>> from transformers import AutoTokenizer
+
+        >>> model = LLaDA2MoeForCausalLM.from_pretrained(PATH_TO_CONVERTED_WEIGHTS)
+        >>> tokenizer = AutoTokenizer.from_pretrained(PATH_TO_CONVERTED_TOKENIZER)
+
+        >>> prompt = "Hey, are you conscious? Can you talk to me?"
+        >>> inputs = tokenizer(prompt, return_tensors="pt")
+
+        >>> # Generate
+        >>> generate_ids = model.generate(inputs.input_ids, max_length=30)
+        >>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
+        "Hey, are you conscious? Can you talk to me?\nI'm not conscious, but I can talk to you."
+        ```"""
+        output_attentions = (
+            output_attentions
+            if output_attentions is not None
+            else self.config.output_attentions
+        )
+        output_hidden_states = (
+            output_hidden_states
+            if output_hidden_states is not None
+            else self.config.output_hidden_states
+        )
+        output_router_logits = (
+            output_router_logits
+            if output_router_logits is not None
+            else self.config.output_router_logits
+        )
+        return_dict = (
+            return_dict if return_dict is not None else self.config.use_return_dict
+        )
+        # decoder outputs consists of (dec_features, layer_state, dec_hidden, dec_attn)
+        outputs = self.model(
+            input_ids=input_ids,
+            attention_mask=attention_mask,
+            position_ids=position_ids,
+            past_key_values=past_key_values,
+            inputs_embeds=inputs_embeds,
+            use_cache=use_cache,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            output_router_logits=output_router_logits,
+            return_dict=return_dict,
+            **kwargs,
+        )
+
+        loss = None
+        aux_loss = None
+        hidden_states = outputs[0]
+
+        logits = self.lm_head(hidden_states)
+        logits = logits.float()
+
+        if labels is not None:
+            # LLaDA2.0 will use same label position logits
+            shift_logits = logits
+            shift_labels = labels
+            # Flatten the tokens
+            loss_fct = CrossEntropyLoss()
+            shift_logits = shift_logits.view(-1, self.config.vocab_size)
+            shift_labels = shift_labels.view(-1)
+            # Enable model parallelism
+            shift_labels = shift_labels.to(shift_logits.device)
+            loss = loss_fct(shift_logits, shift_labels)
+
+        if not return_dict:
+            output = (logits,) + outputs[1:]
+            if output_router_logits:
+                output = (aux_loss,) + output
+            return (loss,) + output if loss is not None else output
+
+        return MoeCausalLMOutputWithPast(
+            loss=loss,
+            aux_loss=aux_loss,
+            logits=logits,
+            past_key_values=outputs.past_key_values,
+            hidden_states=outputs.hidden_states,
+            attentions=outputs.attentions,
+            router_logits=outputs.router_logits,
+        )
+
+    def prepare_inputs_for_generation(
+        self,
+        input_ids,
+        past_key_values=None,
+        attention_mask=None,
+        inputs_embeds=None,
+        token_type_ids=None,
+        **kwargs,
+    ):
+        if past_key_values is not None:
+            if isinstance(past_key_values, Cache):
+                cache_length = past_key_values.get_seq_length()
+                past_length = past_key_values.seen_tokens
+                max_cache_length = (
+                    past_key_values.get_max_length()
+                    if hasattr(past_key_values, "get_max_length")
+                    else past_key_values.get_max_cache_shape()
+                )
+            else:
+                cache_length = past_length = past_key_values[0][0].shape[2]
+                max_cache_length = None
+
+            # Keep only the unprocessed tokens:
+            # 1 - If the length of the attention_mask exceeds the length of input_ids, then we are in a setting where
+            # some of the inputs are exclusivelly passed as part of the cache (e.g. when passing input_embeds as input)
+            if (
+                attention_mask is not None
+                and attention_mask.shape[1] > input_ids.shape[1]
+            ):
+                input_ids = input_ids[:, -(attention_mask.shape[1] - past_length) :]
+            # 2 - If the past_length is smaller than input_ids', then input_ids holds all input tokens. We can discard
+            # input_ids based on the past_length.
+            elif past_length < input_ids.shape[1]:
+                input_ids = input_ids[:, past_length:]
+            # 3 - Otherwise (past_length >= input_ids.shape[1]), let's assume input_ids only has unprocessed tokens.
+
+            # If we are about to go beyond the maximum cache length, we need to crop the input attention mask.
+            if (
+                max_cache_length is not None
+                and attention_mask is not None
+                and cache_length + input_ids.shape[1] > max_cache_length
+            ):
+                attention_mask = attention_mask[:, -max_cache_length:]
+
+        position_ids = kwargs.get("position_ids", None)
+        if attention_mask is not None and position_ids is None:
+            # create position_ids on the fly for batch generation
+            position_ids = attention_mask.long().cumsum(-1) - 1
+            position_ids.masked_fill_(attention_mask == 0, 1)
+            if past_key_values:
+                position_ids = position_ids[:, -input_ids.shape[1] :]
+
+        # if `inputs_embeds` are passed, we only want to use them in the 1st generation step
+        if inputs_embeds is not None and past_key_values is None:
+            model_inputs = {"inputs_embeds": inputs_embeds}
+        else:
+            model_inputs = {"input_ids": input_ids}
+
+        model_inputs.update(
+            {
+                "position_ids": position_ids,
+                "past_key_values": past_key_values,
+                "use_cache": kwargs.get("use_cache"),
+                "attention_mask": attention_mask,
+            }
+        )
+        return model_inputs
+
+    @staticmethod
+    def _reorder_cache(past_key_values, beam_idx):
+        reordered_past = ()
+        for layer_past in past_key_values:
+            reordered_past += (
+                tuple(
+                    past_state.index_select(0, beam_idx.to(past_state.device))
+                    for past_state in layer_past
+                ),
+            )
+        return reordered_past
+
+    @staticmethod
+    def _top_k_logits(logits, k):
+        if k is None or k <= 0:
+            return logits
+        else:
+            values, _ = torch.topk(logits, k)
+            min_values = values[..., -1, None]
+            return torch.where(
+                logits < min_values, torch.full_like(logits, float("-inf")), logits
+            )
+
+    @staticmethod
+    def _top_p_logits(logits, p):
+        if p is None or p >= 1.0:
+            return logits
+        sorted_logits, sorted_indices = torch.sort(logits, descending=True)
+        cumulative_probs = torch.cumsum(F.softmax(sorted_logits, dim=-1), dim=-1)
+        sorted_mask = cumulative_probs > p
+        sorted_mask[..., 1:] = sorted_mask[..., :-1].clone()
+        sorted_mask[..., 0] = False
+        mask_indices = torch.scatter(
+            torch.full_like(logits, False, dtype=torch.bool),
+            -1,
+            sorted_indices,
+            sorted_mask,
+        )
+        return logits.masked_fill(mask_indices, float("-inf"))
+
+    def _sample_with_temperature_topk_topp(
+        self, logits, temperature=1.0, top_k=0, top_p=1.0
+    ):
+        orig_shape = logits.shape[:-1]
+        vocab_size = logits.shape[-1]
+        logits = logits.reshape(-1, vocab_size)
+        if temperature == 0.0:
+            token = torch.argmax(logits, dim=-1, keepdim=True)
+            probs = F.softmax(logits, dim=-1)
+            token_prob = torch.gather(probs, -1, token)
+            return token.view(*orig_shape), token_prob.view(*orig_shape)
+
+        if temperature > 0 and temperature != 1.0:
+            logits = logits / temperature
+        logits = self._top_k_logits(logits, top_k)
+        logits = self._top_p_logits(logits, top_p)
+        probs = F.softmax(logits, dim=-1)
+        token = torch.multinomial(probs, num_samples=1)
+        token_prob = torch.gather(probs, -1, token)
+        return token.view(*orig_shape), token_prob.view(*orig_shape)
+
+    @staticmethod
+    def _get_num_transfer_tokens(block_length, steps):
+        if steps == 0:
+            return torch.tensor([], dtype=torch.int64)
+        base = block_length // steps
+        remainder = block_length % steps
+        num_transfer_tokens = torch.full((steps,), base, dtype=torch.int64)
+        num_transfer_tokens[:remainder] += 1
+        return num_transfer_tokens
+
+    @torch.no_grad()
+    def generate(
+        self,
+        inputs: Optional[torch.Tensor] = None,
+        temperature: float = 0.0,
+        block_length: int = 32,
+        steps: int = 32,
+        gen_length: int = 2048,
+        top_p: Optional[float] = None,
+        top_k: Optional[int] = None,
+        eos_early_stop: bool = False,
+        minimal_topk: int = 1,
+        threshold: float = 0.95,
+        editing_threshold: float = 0.9,
+        max_post_steps: int = 16,
+        eos_id: int = 156892,
+        mask_id: int = 156895,
+        num_to_transfer: int = 1,
+    ):
+        r"""
+        Generates tokens using a block-wise, iterative refinement strategy.
+        This method operates differently from standard autoregressive generation. It first creates a template of the
+        full desired length, filled with a special `mask_id`. It then processes this template in segments (`blocks`)
+        and iteratively "denoises" or "refines" the `mask_id` tokens into actual tokens over a series of `steps` for
+        each block. A custom block-diagonal causal attention mask ensures that generation within a block can attend to
+        all previous blocks but not future ones.
+        <Tip warning={true}>
+        This is a specialized generation method. The quality and speed of the output are highly dependent on the interplay
+        between `block_length`, `steps`, and `threshold`. It aims to achieve faster generation through parallel
+        decoding within blocks, which is a departure from the token-by-token generation of standard `.generate()` methods.
+        </Tip>
+        Parameters:
+            inputs (`torch.Tensor`):
+                The token sequence used as a prompt for the generation.
+            temperature (`float`, *optional*, defaults to 0.0):
+                The value used to module the next token probabilities. A value of 0.0 corresponds to greedy decoding.
+            block_length (`int`, *optional*, defaults to 32):
+                The size of each generation block. The model generates text in parallel within these blocks. This is a
+                key parameter for controlling the granularity of the generation process.
+            steps (`int`, *optional*, defaults to 32):
+                The number of iterative refinement (or "denoising") steps to perform for each block. Within each block,
+                the model will try to replace `mask_id` tokens with real tokens for this many iterations.
+            gen_length (`int`, *optional*, defaults to 2048):
+                The maximum number of tokens to generate, excluding the prompt.
+            top_p (`float`, *optional*):
+                If set to a float value between 0 and 1, only the most probable tokens with probabilities that add up to
+                `top_p` or higher are kept for generation (nucleus sampling).
+            top_k (`int`, *optional*):
+                The number of highest probability vocabulary tokens to keep for top-k-filtering.
+            eos_early_stop (`bool`, *optional*, defaults to `False`):
+                If `True`, generation will stop as soon as a valid End-Of-Sequence token is generated and confirmed,
+                even if `gen_length` has not been reached.
+            minimal_topk (`int`, *optional*, defaults to 1):
+                A parameter used to dynamically adjust the number of refinement `steps`. The effective number of steps
+                is capped at `gen_length // minimal_topk`.
+            threshold (`float`, *optional*, defaults to 0.95):
+                The confidence probability threshold for accepting a sampled token. During each refinement step, a
+                sampled token is only kept if its probability is above this threshold. If not enough tokens meet the
+                threshold, the ones with the highest confidence are chosen.
+            editing_threshold (`float`, *optional*, defaults to 0.5):
+                The confidence threshold for **editing**. Existing tokens (non-masked) are replaced by newly
+                sampled tokens if the model's confidence in the new token exceeds this threshold and the token has changed.
+            max_post_steps (`int`, *optional*, defaults to 16):
+                Number of global refinement iterations after all mask tokens are resolved.
+            eos_id (`int`, *optional*, defaults to 156892):
+                The token ID for the end-of-sequence token. Used for `eos_early_stop`.
+            mask_id (`int`, *optional*, defaults to 156895):
+                The token ID used as a placeholder for tokens that are yet to be generated. This is central to the
+                iterative refinement algorithm.
+        Return:
+            `torch.Tensor`: A string containing the generated token IDs, starting
+            after the prompt and stopping at the first `eos_id` or `gen_length`.
+        """
+
+        steps = min(steps, gen_length // minimal_topk)
+        input_ids = inputs.to(self.device)
+
+        prompt_length = input_ids.shape[1]
+        num_blocks = (prompt_length + gen_length + block_length - 1) // block_length
+        total_length = num_blocks * block_length
+
+        block_mask = torch.tril(torch.ones(num_blocks, num_blocks, device=self.device))
+        block_diffusion_attention_mask = (
+            block_mask.repeat_interleave(block_length, dim=0)
+            .repeat_interleave(block_length, dim=1)
+            .unsqueeze(0)
+            .unsqueeze(0)
+        ).to(torch.bfloat16)
+
+        position_ids = torch.arange(total_length, device=self.device).unsqueeze(0)
+        x = torch.full((1, total_length), mask_id, dtype=torch.long, device=self.device)
+        x[:, :prompt_length] = input_ids.clone()
+
+        prompt_index_full = torch.zeros_like(x, dtype=torch.bool)
+        prompt_index_full[:, :prompt_length] = True
+
+        prefill_blocks = prompt_length // block_length
+
+        for num_block in range(prefill_blocks, num_blocks):
+            current_window_end = (num_block + 1) * block_length
+            cur_x = x[:, :current_window_end]
+            cur_attn_mask = block_diffusion_attention_mask[
+                :, :, :current_window_end, :current_window_end
+            ]
+            cur_position_ids = position_ids[:, :current_window_end]
+
+            block_start_pos = num_block * block_length
+
+            post_steps = 0
+            while True:
+                old_block_tokens = cur_x[:, -block_length:].clone()
+                active_block_mask = cur_x[:, -block_length:] == mask_id
+                if torch.any(active_block_mask) == False:
+                    post_steps += 1
+                if post_steps > max_post_steps:
+                    break
+                prompt_mask_in_block = torch.zeros(
+                    block_length, dtype=torch.bool, device=self.device
+                )
+                if block_start_pos < prompt_length:
+                    prompt_end_in_block = min(
+                        prompt_length - block_start_pos, block_length
+                    )
+                    prompt_mask_in_block[:prompt_end_in_block] = True
+
+                outputs = self.forward(
+                    cur_x,
+                    attention_mask=cur_attn_mask,
+                    position_ids=cur_position_ids,
+                    output_attentions=True,
+                )
+                logits = outputs.logits
+
+                active_logits = logits[:, -block_length:, :]
+                x0, x0_p = self._sample_with_temperature_topk_topp(
+                    active_logits, temperature=temperature, top_k=top_k, top_p=top_p
+                )
+                mask_transfer_index = torch.zeros_like(x0, dtype=torch.bool)
+                if active_block_mask.sum() > 0:
+                    mask_confidence = torch.where(active_block_mask, x0_p, -torch.inf)
+                    high_conf_mask = (
+                        mask_confidence[0] > threshold
+                    ) & active_block_mask[0]
+                    num_high_confidence = high_conf_mask.sum().item()
+
+                    if num_high_confidence >= num_to_transfer:
+                        mask_transfer_index[0] = high_conf_mask
+                    else:
+                        num_available = active_block_mask.sum().item()
+                        if num_available > 0:
+                            _, idx = torch.topk(
+                                mask_confidence[0],
+                                k=min(num_to_transfer, num_available),
+                            )
+                            mask_transfer_index[0, idx] = True
+
+                editing_transfer_index = torch.zeros_like(x0, dtype=torch.bool)
+                non_mask_positions = ~active_block_mask
+                non_prompt_positions = ~prompt_mask_in_block
+                editable_positions = non_mask_positions & non_prompt_positions[None, :]
+                editing_confidence = torch.where(editable_positions, x0_p, -torch.inf)
+                high_conf_editing = (
+                    editing_confidence[0] > editing_threshold
+                ) & editable_positions[0]
+
+                token_changed = x0[0] != old_block_tokens[0]
+                editing_transfer_index[0] = high_conf_editing & token_changed
+                final_transfer_index = mask_transfer_index | editing_transfer_index
+
+                if final_transfer_index.any():
+                    cur_x[:, -block_length:][final_transfer_index] = x0[
+                        final_transfer_index
+                    ]
+
+                if active_block_mask.sum() == 0 and not editing_transfer_index.any():
+                    break
+
+            x[:, :current_window_end] = cur_x
+            if eos_early_stop:
+                generated_part = x[0, prompt_length:current_window_end]
+                if (generated_part == mask_id).sum() == 0:
+                    eos_positions = (generated_part == eos_id).nonzero(as_tuple=True)[0]
+                    if len(eos_positions) > 0:
+                        break
+
+        generated_answer = x[:, : prompt_length + gen_length]
+        mask_positions = (generated_answer[0][input_ids.shape[1] :] == eos_id).nonzero(
+            as_tuple=True
+        )[0]
+        if len(mask_positions) > 0:
+            first_mask_position = mask_positions[0].item()
+        else:
+            first_mask_position = gen_length
+
+        return generated_answer[
+            :, input_ids.shape[1] : input_ids.shape[1] + first_mask_position + 1
+        ]

special_tokens_map.json

@@ -0,0 +1,8 @@
+{
+  "bos_token": "<|startoftext|>",
+  "cls_token": "[CLS]",
+  "eos_token": "<|endoftext|>",
+  "gmask_token": "[gMASK]",
+  "pad_token": "<|endoftext|>",
+  "mask_token": "<|mask|>"
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:878eb900ea4a42da7e54bc18ad38352bec005bbfd81bdf76a5ecf1aad1093a6c
+size 12205801

tokenizer_config.json

@@ -0,0 +1,19 @@
+{
+  "backend": "tokenizers",
+  "bos_token": "<|startoftext|>",
+  "clean_up_tokenization_spaces": false,
+  "cls_token": "[CLS]",
+  "eos_token": "<|endoftext|>",
+  "fast_tokenizer": true,
+  "gmask_token": "[gMASK]",
+  "is_local": true,
+  "local_files_only": false,
+  "mask_token": "<|mask|>",
+  "merges_file": null,
+  "model_max_length": 32768,
+  "model_specific_special_tokens": {
+    "gmask_token": "[gMASK]"
+  },
+  "pad_token": "<|endoftext|>",
+  "tokenizer_class": "TokenizersBackend"
+}