---

functionary-small-v3.2-GGUF

Original Model

Run with LlamaEdge

• LlamaEdge version: v0.14.10 and above
• Prompt template
  • Prompt type: functionary-32
  • Prompt string
    text

    1<|start_header_id|>system<|end_header_id|> 2 3You are capable of executing available function(s) if required. 4Only execute function(s) when absolutely necessary. 5Ask for the required input to:recipient==all 6Use JSON for function arguments. 7Respond in this format: 8>>>${recipient} 9${content} 10Available functions: 11// Supported function definitions that should be called when necessary. 12namespace functions { 13 14 // Get the current weather 15 type get_current_weather = (_: { 16 17 // The city and state, e.g. San Francisco, CA 18 location: string, 19 20 }) => any; 21 22 23} // namespace functions<|eot_id|><|start_header_id|>user<|end_header_id|> 24 25What is the weather like in Beijing today?<|eot_id|><|start_header_id|>assistant<|end_header_id|>
• Context size: 128000
• Run as LlamaEdge service
  bash

  1wasmedge --dir .:. --nn-preload default:GGML:AUTO:functionary-small-v3.2-Q5_K_M.gguf \ 2 llama-api-server.wasm \ 3 --model-name functionary-small-v3.2 \ 4 --prompt-template functionary-32 \ 5 --ctx-size 128000
• Run as LlamaEdge command app
  bash

  1wasmedge --dir .:. --nn-preload default:GGML:AUTO:functionary-small-v3.2-Q5_K_M.gguf \ 2 llama-chat.wasm \ 3 --prompt-template functionary-32 \ 4 --ctx-size 128000

Quantized GGUF Models