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

chat_template.jinja

@@ -0,0 +1,159 @@
+{# ---------------------------------------------------------------------- #}
+{# ƛƬ Default setup and flags                                             #}
+{# ---------------------------------------------------------------------- #}
+{%- set messages = messages or [] -%}
+{%- set tools = tools or [] -%}
+{%- set add_generation_prompt = add_generation_prompt or false -%}
+{%- set available_tool_string, add_tool_id = '', true -%}
+{%- set add_thoughts = false -%}            {# whether to include <thinking> reasoning blocks #}
+{%- set add_generation_prompt = true -%}      {# whether to emit reasoning starter before assistant response #}
+{# Optional token placeholders (safe defaults) #}
+{%- set bos_token = bos_token if (bos_token is defined) else '' -%}
+{%- set eos_token = eos_token if (eos_token is defined) else '' -%}
+{# ---------------------------------------------------------------------- #}
+{# Core reasoning prompt and assistant reasoning prefix                 #}
+{# ---------------------------------------------------------------------- #}
+{%- set reasoning_prompt = 
+    'You are a thoughtful, systematic AI assistant from ServiceNow Language Models (SLAM) lab. '
+    'Analyze each question carefully, present your reasoning step-by-step, then provide the final '
+    'response after the marker [BEGIN FINAL RESPONSE].'
+-%}
+{%- set reasoning_asst_turn_start = 'Here are my reasoning steps:\n' -%}
+{# ---------------------------------------------------------------------- #}
+{# Tool list and tool call output format                                  #}
+{# ---------------------------------------------------------------------- #}
+{%- if tools is not none and tools|length > 0 -%}
+    {%- set available_tool_string -%}
+        You are provided with function signatures within <available_tools></available_tools> XML tags.
+        You may call one or more functions to assist with the user query.
+        Don't make assumptions about the arguments. You should infer the argument values from previous
+        user responses and the system message.
+        Here are the available tools: 
+        <available_tools>
+        {% for tool in tools %}{{ tool|string }}{% endfor %}
+        
+        </available_tools>.
+
+        Return all function calls as a list of JSON objects within <tool_calls></tool_calls> XML tags.
+        Each JSON object should contain a function name and arguments as follows:
+        <tool_calls>[
+            {"name": <function-name-1>, "arguments": <args-dict-1>},
+            {"name": <function-name-2>, "arguments": <args-dict-2>},
+            ...
+        ]</tool_calls>
+    {%- endset -%}
+{%- endif -%}
+{# ---------------------------------------------------------------------- #}
+{# Start system block if first message is not system                      #}
+{# ---------------------------------------------------------------------- #}
+{%- if messages|length > 0 and messages[0]['role'] != 'system' -%}
+    {%- if tools is not none and tools|length > 0 -%}
+        {{ bos_token + '<|begin_system|>\n' + reasoning_prompt + '\n' + available_tool_string + '\n' }}
+    {%- else -%}
+        {{ bos_token + '<|begin_system|>\n' + reasoning_prompt + '\n' }}
+    {%- endif -%}
+{%- endif -%}
+{# ---------------------------------------------------------------------- #}
+{# Iterate through messages                                             #}
+{# ---------------------------------------------------------------------- #}
+{%- for message in messages -%}
+
+    {# ---------------- USER MESSAGE ---------------- #}
+    {%- if message['role'] == 'user' -%}
+        {{ '<|begin_user|>\n' }}
+        {%- if message['content'] is not string -%}
+            {%- for chunk in message['content'] -%}
+                {%- if chunk['type'] == 'text' -%}
+                    {{ chunk['text'] }}
+                {%- elif chunk['type'] in ['image', 'image_url'] -%}
+                    {{ '[IMG]' }}
+                {%- else -%}
+                    {{ raise_exception('Unrecognized content type!') }}
+                {%- endif -%}
+            {%- endfor -%}
+        {%- else -%}
+            {{ message['content'] }}
+        {%- endif -%}
+
+    {# ---------------- SYSTEM MESSAGE ---------------- #}
+    {%- elif message['role'] == 'system' -%}
+        {%- if message['content'] is not none and message['content']|length > 0 -%}
+            {%- if message['content'] is string -%}
+                {%- set system_message = message['content'] -%}
+            {%- else -%}
+                {%- set system_message = message['content'][0]['text'] -%}
+            {%- endif -%}
+        {%- else -%}
+            {%- set system_message = '' -%}
+        {%- endif -%}
+
+        {%- if tools is not none and tools|length > 0 -%}
+            {{ bos_token + '<|begin_system|>\n' + reasoning_prompt + '\n' + system_message + '\n' + available_tool_string + '\n' }}
+        {%- else -%}
+            {{ bos_token + '<|begin_system|>\n' + reasoning_prompt + '\n' + system_message + '\n' }}
+        {%- endif -%}
+
+    {# ---------------- ASSISTANT MESSAGE ---------------- #}
+    {%- elif message['role'] == 'assistant' -%}
+        {%- if loop.last -%}
+            {%- set add_tool_id = false -%}
+        {%- endif -%}
+
+        {{ '\n<|begin_assistant|>\n' }}
+
+        {%- if add_thoughts and 'thought' in message and message['thought'] is not none -%}
+            <thinking>{{ message['thought'] }}</thinking>
+        {%- endif -%}
+
+        {%- if message['content'] is not none and message['content']|length > 0 -%}
+            {%- if message['content'] is not string -%}
+                {{ message['content'][0]['text'] }}
+            {%- else -%}
+                {{ message['content'] }}
+            {%- endif -%}
+        {%- elif message['chosen'] is not none and message['chosen']|length > 0 -%}
+            {{ message['chosen'][0] }}
+        {%- endif -%}
+
+        {# Tool call output #}
+        {%- if message['tool_calls'] is not none and message['tool_calls']|length > 0 -%}
+            {{ '\n<tool_calls>[' }}
+            {%- for tool_call in message['tool_calls'] -%}
+                {{ '{"name": "' + tool_call['function']['name'] + '", "arguments": ' + tool_call['function']['arguments']|string }}
+                {%- if add_tool_id == true and 'id' in tool_call -%}
+                    {{ ', "id": "' + tool_call['id'] + '"' }}
+                {%- endif -%}
+                {{ '}' }}
+                {%- if not loop.last -%}{{ ', ' }}{%- endif -%}
+            {%- endfor -%}
+            {{ ']</tool_calls>' }}
+        {%- endif -%}
+
+        {%- if not loop.last or training_prompt -%}
+            {{ '\n<|end|>\n' }}
+        {%- endif -%}
+
+    {# ---------------- TOOL RESULT MESSAGE ---------------- #}
+    {%- elif message['role'] == 'tool' -%}
+        {%- if message['content'] is string -%}
+            {%- set tool_message = message['content'] -%}
+        {%- else -%}
+            {%- set tool_message = message['content'][0]['text'] -%}
+        {%- endif -%}
+        {{ '<|begin_tool_result|>\n' + tool_message|string + '\n' }}
+
+    {# ---------------- CONTENT MESSAGE ---------------- #}
+    {%- elif message['role'] == 'content' -%}
+        {%- if message['content'] is not string -%}
+            {{ '<|begin_content|>\n' + message['content'][0]['text'] + '\n' }}
+        {%- else -%}
+            {{ '<|begin_content|>\n' + message['content'] + '\n' }}
+        {%- endif -%}
+    {%- endif -%}
+
+    {# ---------------- REASONING PROMPT BEFORE NEXT ASSISTANT ---------------- #}
+    {%- if loop.last and add_generation_prompt and message['role'] != 'assistant' -%}
+        {{ '\n<|begin_assistant|>\n' + reasoning_asst_turn_start }}
+    {%- endif -%} 
+
+{%- endfor -%}

config.json

@@ -0,0 +1,151 @@
+{
+  "architectures": [
+    "Apriel2ForConditionalGeneration"
+  ],
+  "auto_map": {
+    "AutoConfig": "configuration_apriel2.Apriel2Config",
+    "AutoModel": "modeling_apriel2.Apriel2Model",
+    "AutoModelForImageTextToText": "modeling_apriel2.Apriel2ForConditionalGeneration"
+  },
+  "bos_token_id": 1,
+  "decoder": {
+    "block": {
+      "mixer": {
+        "main_mixer_name": "attention",
+        "mixers": {
+          "attention": {
+            "add_linear_biases": false,
+            "head_groups": 8,
+            "head_size": 128,
+            "heads": 32,
+            "rotary": {
+              "theta": 1000000000.0,
+              "type": "mistral_1d"
+            },
+            "type": "attention",
+            "window_size": null
+          },
+          "gdn": {
+            "convolution_layer": {
+              "kernel_size": 4
+            },
+            "key_head_dim": 128,
+            "key_heads": 8,
+            "type": "gdn",
+            "value_head_dim": 128,
+            "value_heads": 32
+          },
+          "kda": {
+            "convolution_layer": {
+              "kernel_size": 4
+            },
+            "head_dim": 128,
+            "heads": 32,
+            "normalization": {
+              "epsilon": 1e-05
+            },
+            "type": "kda"
+          },
+          "sliding_window": {
+            "add_linear_biases": false,
+            "head_groups": 8,
+            "head_size": 128,
+            "heads": 32,
+            "rotary": {
+              "theta": 1000000000.0,
+              "type": "mistral_1d"
+            },
+            "type": "attention",
+            "window_size": 4096
+          }
+        },
+        "sampling_strategy": "uniform",
+        "type": "stochastic"
+      },
+      "mlp": {
+        "activation": "silu",
+        "add_linear_biases": false,
+        "gated": true,
+        "intermediate_size": 14336,
+        "type": "mlp"
+      },
+      "normalization": {
+        "epsilon": 1e-05,
+        "type": "rms_norm"
+      }
+    },
+    "num_blocks": 48,
+    "type": "fixed"
+  },
+  "embeddings": {
+    "max_position_embeddings": 120000
+  },
+  "eos_token_id": 2,
+  "head": {
+    "normalization": {
+      "epsilon": 1e-05,
+      "type": "rms_norm"
+    }
+  },
+  "hidden_size": 5120,
+  "image_token_index": 10,
+  "model_type": "apriel2",
+  "tie_word_embeddings": false,
+  "transformers_version": "4.57.3",
+  "use_cache": true,
+  "vision_encoder": {
+    "adapter": {
+      "activation": "gelu_pytorch_tanh",
+      "add_linear_biases": true,
+      "gated": false,
+      "intermediate_size": 5120,
+      "type": "mlp"
+    },
+    "embeddings": {
+      "input_channels": 3,
+      "normalization": {
+        "epsilon": 1e-05,
+        "type": "rms_norm"
+      },
+      "patch_height": 16,
+      "patch_width": 16
+    },
+    "encoder": {
+      "block": {
+        "mixer": {
+          "add_linear_biases": false,
+          "causal": false,
+          "cross_document_attention": false,
+          "head_groups": 16,
+          "head_size": 64,
+          "heads": 16,
+          "rotary": {
+            "max_image_size": 1024,
+            "patch_size": 16,
+            "theta": 10000.0,
+            "type": "pixtral_2d"
+          },
+          "type": "attention"
+        },
+        "mlp": {
+          "activation": "silu",
+          "add_linear_biases": false,
+          "gated": true,
+          "intermediate_size": 4096,
+          "type": "mlp"
+        },
+        "normalization": {
+          "epsilon": 1e-05,
+          "type": "rms_norm"
+        }
+      },
+      "num_blocks": 24,
+      "type": "fixed"
+    },
+    "hidden_size": 1024,
+    "num_channels": 3,
+    "num_hidden_layers": 24,
+    "patch_size": 16
+  },
+  "vocab_size": 131072
+}

configuration_apriel2.py

@@ -0,0 +1,144 @@
+"""Apriel2 HuggingFace configuration."""
+
+import logging
+from typing import Optional
+
+from transformers import PretrainedConfig
+
+logger = logging.getLogger(__name__)
+
+
+class Apriel2TextConfig(PretrainedConfig):
+    model_type = "apriel2_text"
+
+    def __init__(
+        self,
+        hidden_size: int = 4096,
+        vocab_size: int = 32000,
+        decoder: Optional[dict] = None,
+        embeddings: Optional[dict] = None,
+        head: Optional[dict] = None,
+        tie_word_embeddings: bool = False,
+        bos_token_id: int = 1,
+        eos_token_id: int = 2,
+        pad_token_id: Optional[int] = None,
+        use_cache: bool = True,
+        **kwargs,
+    ):
+        self.hidden_size = hidden_size
+        self.vocab_size = vocab_size
+        self.use_cache = use_cache
+
+        self.decoder = decoder or self._default_decoder_config()
+        self.embeddings = embeddings or self._default_embeddings_config()
+        self.head = head or self._default_head_config()
+
+        super().__init__(
+            bos_token_id=bos_token_id,
+            eos_token_id=eos_token_id,
+            pad_token_id=pad_token_id,
+            tie_word_embeddings=tie_word_embeddings,
+            **kwargs,
+        )
+
+    def _default_decoder_config(self) -> dict:
+        return {
+            "type": "fixed",
+            "num_blocks": 32,
+            "block": {
+                "mixer": {
+                    "type": "attention",
+                    "heads": 32,
+                    "head_groups": 32,
+                    "head_size": self.hidden_size // 32,
+                    "rotary": {"type": "default", "theta": 10000.0},
+                    "add_linear_biases": False,
+                },
+                "mlp": {
+                    "type": "mlp",
+                    "intermediate_size": self.hidden_size * 4,
+                    "activation": "silu",
+                    "gated": True,
+                    "add_linear_biases": False,
+                },
+                "normalization": {"type": "rms_norm", "epsilon": 1e-5},
+            },
+        }
+
+    def _default_embeddings_config(self) -> dict:
+        return {
+            "max_position_embeddings": 2048,
+        }
+
+    def _default_head_config(self) -> dict:
+        return {
+            "normalization": {"type": "rms_norm", "epsilon": 1e-5},
+        }
+
+    def get_text_config(self, decoder: bool = False):
+        return self
+
+    def get_block_name(self, layer_idx: int) -> str:
+        decoder_type = self.decoder.get("type", "fixed")
+
+        if decoder_type == "fixed":
+            return "block"
+        elif decoder_type == "pattern":
+            pattern = self.decoder.get("pattern", [])
+            if not pattern:
+                raise ValueError("Pattern decoder requires 'pattern' field")
+            return pattern[layer_idx % len(pattern)]
+        else:
+            raise ValueError(f"Unknown decoder type: {decoder_type}")
+
+    def get_block_config(self, layer_idx: int) -> dict:
+        decoder_type = self.decoder.get("type", "fixed")
+
+        if decoder_type == "fixed":
+            return self.decoder.get("block", {})
+        elif decoder_type == "pattern":
+            blocks = self.decoder.get("blocks", {})
+            pattern = self.decoder.get("pattern", [])
+            if not blocks or not pattern:
+                raise ValueError("Pattern decoder requires 'blocks' and 'pattern' fields")
+            block_name = pattern[layer_idx % len(pattern)]
+            return blocks[block_name]
+        else:
+            raise ValueError(f"Unknown decoder type: {decoder_type}")
+
+
+class Apriel2Config(Apriel2TextConfig):
+    model_type = "apriel2"
+
+    def __init__(
+        self,
+        hidden_size: int = 4096,
+        vocab_size: int = 32000,
+        decoder: Optional[dict] = None,
+        embeddings: Optional[dict] = None,
+        head: Optional[dict] = None,
+        vision_encoder: Optional[dict] = None,
+        image_token_index: Optional[int] = None,
+        tie_word_embeddings: bool = False,
+        bos_token_id: int = 1,
+        eos_token_id: int = 2,
+        pad_token_id: Optional[int] = None,
+        use_cache: bool = True,
+        **kwargs,
+    ):
+        super().__init__(
+            hidden_size=hidden_size,
+            vocab_size=vocab_size,
+            decoder=decoder,
+            embeddings=embeddings,
+            head=head,
+            tie_word_embeddings=tie_word_embeddings,
+            bos_token_id=bos_token_id,
+            eos_token_id=eos_token_id,
+            pad_token_id=pad_token_id,
+            use_cache=use_cache,
+            **kwargs,
+        )
+
+        self.vision_encoder = vision_encoder
+        self.image_token_index = image_token_index

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "transformers_version": "4.49.0"
+}

model_0.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8d8388e1c0436b65a78bddb703043fa6436b64ab8c16fff8c447e5387137b8de
+size 17425048856

model_1.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57e36c17b1e3117255fab5fc00715dfba4ea71b94d2ff4baeae148b489be01eb
+size 17470320128

model_2.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0af1648b2f47fa69cc71867d88fb05fc28c949e3a8c84043acd77caa3d4a6a9f
+size 17470320304

model_3.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aac25febd9dbc7f067d264c08ad36216501e91c3d4c3a5e223cb32f8fd6e89f7
+size 17470320304

model_4.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9ef25d09db12a0c5b9e8cb2d63721ea60b6b9c6c0692df31f19b32286f042095
+size 17470320304

model_5.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d7ccbcb5ec330e2ddc9fb19a1efebf610d60940acb47258c9f280951d9ae63ab
+size 12977332728

preprocessor_config.json

@@ -0,0 +1,27 @@
+{
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_processor_type": "PixtralImageProcessor",
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "patch_size": {
+    "height": 16,
+    "width": 16
+  },
+  "processor_class": "PixtralProcessor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "longest_edge": 1024
+  }
+}

special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:50c4196bd3d61abf4a6f9d116435140b8ac0606e15eb4d02e235f9036257dc3e
+size 17077327