---

Meta-DeepHermes-3-Llama-3-8B-Preview-GGUF

Original Model

NousResearch/DeepHermes-3-Llama-3-8B-Preview

Run with LlamaEdge

• LlamaEdge version: v0.16.8 and above

• Prompt template

  IMPORTANT: To toggle REASONING ON, you must use the following system prompt:

  You are a deep thinking AI, you may use extremely long chains of thought to deeply consider the problem and deliberate with yourself via systematic reasoning processes to help come to a correct solution prior to answering. You should enclose your thoughts and internal monologue inside <think> </think> tags, and then provide your solution or response to the problem.
  

  • Prompt type for chat: llama-3-chat

    • Prompt string

      <|begin_of_text|><|start_header_id|>system<|end_header_id|>
      
      {{ system_prompt }}<|eot_id|><|start_header_id|>user<|end_header_id|>
      
      {{ user_message_1 }}<|eot_id|><|start_header_id|>assistant<|end_header_id|>
      
      {{ model_answer_1 }}<|eot_id|><|start_header_id|>user<|end_header_id|>
      
      {{ user_message_2 }}<|eot_id|><|start_header_id|>assistant<|end_header_id|>
      

  • Prompt type for tool use: llama-3-tool

    • Prompt string

      <|begin_of_text|><|start_header_id|>system<|end_header_id|>
      
      {system_message}<|eot_id|><|start_header_id|>user<|end_header_id|>
      
      Given the following functions, please respond with a JSON for a function call with its proper arguments that best answers the given prompt.
      
      Respond in the format {"name": function name, "parameters": dictionary of argument name and its value}. Do not use variables.
      
      [{"type":"function","function":{"name":"get_current_weather","description":"Get the current weather in a given location","parameters":{"type":"object","properties":{"location":{"type":"string","description":"The city and state, e.g. San Francisco, CA"},"unit":{"type":"string","description":"The temperature unit to use. Infer this from the users location.","enum":["celsius","fahrenheit"]}},"required":["location","unit"]}}}]
      
      Question: {user_message}<|eot_id|><|start_header_id|>assistant<|end_header_id|>
      

• Context size: 128000

• Run as LlamaEdge service

  • Chat

    wasmedge --dir .:. --nn-preload default:GGML:AUTO:DeepHermes-3-Llama-3-8B-Preview-Q5_K_M.gguf \
      llama-api-server.wasm \
      --prompt-template llama-3-chat \
      --ctx-size 128000 \
      --model-name DeepHermes-3-Llama-3-8B-Preview
    

  • Tool use

    wasmedge --dir .:. --nn-preload default:GGML:AUTO:DeepHermes-3-Llama-3-8B-Preview-Q5_K_M.gguf \
      llama-api-server.wasm \
      --prompt-template llama-3-tool \
      --ctx-size 128000 \
      --model-name DeepHermes-3-Llama-3-8B-Preview
    

• Run as LlamaEdge command app

  wasmedge --dir .:. --nn-preload default:GGML:AUTO:DeepHermes-3-Llama-3-8B-Preview-Q5_K_M.gguf \
    llama-chat.wasm \
    --prompt-template llama-3-chat \
    --ctx-size 128000
  

Quantized GGUF Models

Name	Quant method	Bits	Size	Use case
DeepHermes-3-Llama-3-8B-Preview-Q2_K.gguf	Q2_K	2	3.18 GB	smallest, significant quality loss - not recommended for most purposes
DeepHermes-3-Llama-3-8B-Preview-Q3_K_L.gguf	Q3_K_L	3	4.32 GB	small, substantial quality loss
DeepHermes-3-Llama-3-8B-Preview-Q3_K_M.gguf	Q3_K_M	3	4.02 GB	very small, high quality loss
DeepHermes-3-Llama-3-8B-Preview-Q3_K_S.gguf	Q3_K_S	3	3.66 GB	very small, high quality loss
DeepHermes-3-Llama-3-8B-Preview-Q4_0.gguf	Q4_0	4	4.66 GB	legacy; small, very high quality loss - prefer using Q3_K_M
DeepHermes-3-Llama-3-8B-Preview-Q4_K_M.gguf	Q4_K_M	4	4.92 GB	medium, balanced quality - recommended
DeepHermes-3-Llama-3-8B-Preview-Q4_K_S.gguf	Q4_K_S	4	4.69 GB	small, greater quality loss
DeepHermes-3-Llama-3-8B-Preview-Q5_0.gguf	Q5_0	5	5.6 GB	legacy; medium, balanced quality - prefer using Q4_K_M
DeepHermes-3-Llama-3-8B-Preview-Q5_K_M.gguf	Q5_K_M	5	5.73 GB	large, very low quality loss - recommended
DeepHermes-3-Llama-3-8B-Preview-Q5_K_S.gguf	Q5_K_S	5	5.6 GB	large, low quality loss - recommended
DeepHermes-3-Llama-3-8B-Preview-Q6_K.gguf	Q6_K	6	6.6 GB	very large, extremely low quality loss
DeepHermes-3-Llama-3-8B-Preview-Q8_0.gguf	Q8_0	8	8.54 GB	very large, extremely low quality loss - not recommended
DeepHermes-3-Llama-3-8B-Preview-f16.gguf	f16	16	16.1 GB

Quantized with llama.cpp b4743.