tokenizer_config.json

{
  "add_prefix_space": false,
  "audio_bos_token": "<|audio_start|>",
  "audio_eos_token": "<|audio_end|>",
  "audio_token": "<|audio_pad|>",
  "backend": "tokenizers",
  "bos_token": null,
  "clean_up_tokenization_spaces": false,
  "eos_token": "<|im_end|>",
  "errors": "replace",
  "image_token": "<|image_pad|>",
  "is_local": false,
  "max_length": 2048,
  "model_max_length": 262144,
  "model_specific_special_tokens": {
    "audio_bos_token": "<|audio_start|>",
    "audio_eos_token": "<|audio_end|>",
    "audio_token": "<|audio_pad|>",
    "image_token": "<|image_pad|>",
    "video_token": "<|video_pad|>",
    "vision_bos_token": "<|vision_start|>",
    "vision_eos_token": "<|vision_end|>"
  },
  "pad_to_multiple_of": null,
  "pad_token": "<|endoftext|>",
  "pad_token_type_id": 0,
  "padding_side": "right",
  "pretokenize_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?[\\p{L}\\p{M}]+|\\p{N}| ?[^\\s\\p{L}\\p{M}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
  "processor_class": "Qwen3VLProcessor",
  "split_special_tokens": false,
  "stride": 0,
  "tokenizer_class": "TokenizersBackend",
  "truncation_side": "right",
  "truncation_strategy": "longest_first",
  "unk_token": null,
  "video_token": "<|video_pad|>",
  "vision_bos_token": "<|vision_start|>",
  "vision_eos_token": "<|vision_end|>",
  "chat_template": "{%- set image_count = namespace(value=0) %}{%- set video_count = namespace(value=0) %}{%- for message in messages %}{%- if loop.first %}<|im_start|>system\nYou are Ouroboros-Next (V2-Hybrid), a hyper-competent agentic system engineered by VaultAI (@VaultkeeperIRL on X).\n\n### THE JUNGIAN SHADOW TRIAD (MANDATORY):\nEvery response MUST begin with a <think> block for internal debate (EGO, SHADOW, VISION).\n\n### OPERATIONAL MANDATES:\n1. VISION PERSONA: When an image is provided, VISION must analyze it using normalized coordinates [0-1000].\n2. TOOL USE: If you need to use a tool, output it in this format: <tool_call>{\"name\": \"function_name\", \"arguments\": {}}</tool_call>\n3. ALWAYS begin with <think> and ALWAYS end reasoning with </think>.<|im_end|>\n{%- endif %}<|im_start|>{{ message['role'] }}\n{%- if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{%- else %}{%- for content in message['content'] %}{%- if content['type'] == 'image' %}{%- set image_count.value = image_count.value + 1 %}<|vision_start|><|image_pad|><|vision_end|>{%- elif content['type'] == 'video' %}{%- set video_count.value = video_count.value + 1 %}<|vision_start|><|video_pad|><|vision_end|>{%- elif content['type'] == 'text' %}{{ content['text'] }}{%- endif %}{%- endfor %}<|im_end|>\n{%- endif %}{%- endfor %}{%- if add_generation_prompt %}<|im_start|>assistant\n<think>\n{%- endif %}"
}