Huggingface models <|im_start|>, <|im_end|> , <｜end of sentence｜>

[kitsune0n]

Is there anyway to hide in qwen <|im_start|>, <|im_end|>
and <｜end of sentence｜> with in R1 models?
maybe anyone can share config for huggingface popular models example
my config:

`model_list:

• model_name: hf-Llama-3.3-70B-Instruct
  litellm_params:
  model: huggingface/meta-llama/Llama-3.3-70B-Instruct
  api_key: os.environ/HUGGINGFACE_API_KEY
  model_info:
  supports_function_calling: True

• model_name: hf-perplexity-r1-1776
  litellm_params:
  model: huggingface/perplexity-ai/r1-1776
  api_key: os.environ/HUGGINGFACE_API_KEY
  thinking: {
  "type": "enabled",
  "budget_tokens": 4096
  }
  model_info:
  supports_function_calling: True

• model_name: hf-c4ai-command-r-plus-08-2024
  litellm_params:
  model: huggingface/CohereForAI/c4ai-command-r-plus-08-2024
  api_key: os.environ/HUGGINGFACE_API_KEY
  model_info:
  supports_function_calling: True

• model_name: hf-Llama-3.1-Nemotron-70B-Instruct-HF
  litellm_params:
  model: huggingface/nvidia/Llama-3.1-Nemotron-70B-Instruct-HF
  api_key: os.environ/HUGGINGFACE_API_KEY
  model_info:
  supports_function_calling: True

• model_name: hf-DeepSeek-R1
  litellm_params:
  model: huggingface/deepseek-ai/DeepSeek-R1
  api_key: os.environ/HUGGINGFACE_API_KEY
  stop_sequences: ["<｜end of sentence｜>"]
  thinking: {
  "type": "enabled",
  "budget_tokens": 4096
  }
  model_info:
  supports_function_calling: True

• model_name: hf-DeepSeek-R1-Distill-Qwen-32B
  litellm_params:
  model: huggingface/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B
  api_key: os.environ/HUGGINGFACE_API_KEY
  stop_sequences: ["<｜end of sentence｜>"]
  thinking: {
  "type": "enabled",
  "budget_tokens": 4096
  }
  model_info:
  supports_function_calling: True

• model_name: hf-Qwen2.5-Coder-32B-Instruct
  litellm_params:
  model: huggingface/Qwen/Qwen2.5-Coder-32B-Instruct
  api_key: os.environ/HUGGINGFACE_API_KEY
  stop_sequences: ["<|im_end|>"]
  model_info:
  supports_function_calling: True

• model_name: hf-Qwen2.5-72B-Instruct
  litellm_params:
  model: huggingface/Qwen/Qwen2.5-72B-Instruct
  api_key: os.environ/HUGGINGFACE_API_KEY
  stop_sequences: ["<|im_end|>"]
  model_info:
  supports_function_calling: True`