other-emerald•15mo ago

perflexity

Hi, I'm trying to utilize the Perplexity API with Voiceflow, but I keep receiving a 400 Bad Request error. Does anyone know how to resolve this? Thanks.

10 Replies

Tico | Voiceflow Assistant•15mo ago

Voiceflow provides a platform for creating voice and chat assistants. It offers various functionalities such as interacting with third-party APIs or your own backend services, making network requests, and processing responses. Here's an example of how to make a GET request and process the response:

export default async function main(args) {
  const response = await fetch('https://cat-fact.herokuapp.com/facts');
  const responseBody = response.json();
  // Accessing the response body
  // ... process responseBody
}

export default async function main(args) {
  const response = await fetch('https://cat-fact.herokuapp.com/facts');
  const responseBody = response.json();
  // Accessing the response body
  // ... process responseBody
}

You can also create traces for each item you want to include in the assistant's response:

return next({ path: 'success', trace: facts.map(text => ({ type: 'text', payload: { message: text } })) });

return next({ path: 'success', trace: facts.map(text => ({ type: 'text', payload: { message: text } })) });

Voiceflow also supports various trace types, including text, visual content, cards, and more. For example, to perform a POST request, you can use the following code:

await fetch('YOUR-NGROK-URL-HERE', {
  method: 'POST',
  headers: { 'Content-Type': 'application/json' },
  body: JSON.stringify({ name, age })
});

await fetch('YOUR-NGROK-URL-HERE', {
  method: 'POST',
  headers: { 'Content-Type': 'application/json' },
  body: JSON.stringify({ name, age })
});

For more detailed guidance on making API requests in Voiceflow, you can refer to our Voiceflow Functions Documentation.

W. Williams (SFT)•15mo ago

https://www.voiceflow.com/function/send-question-to-perplexity

Send Question to Perplexity | Voiceflow Functions

Send a question to Perplexity's AI models. This supports pplx-7b-online and pplx-70b-online.

W. Williams (SFT)•15mo ago

There is a function to do it super easily

genetic-orange•15mo ago

still gets a bad request

W. Williams (SFT)•15mo ago

in the perplexity function?

genetic-orange•15mo ago

yeah it's a real mess it's very different from the other chat completion functions

W. Williams (SFT)•15mo ago

I haven't used it but it couldn't be that different. Let me have a look at the API docs and see if I can through something together real quick.

genetic-orange•15mo ago

the endpoint works exactly like together.ai or mistral or anyscale i think i'm working on it myself, just tweaking the function i used for anyscale yeah it does look like it's a function error i'm gonna try to get a test perplexity api key to confirm it

export default async function main(args) {
  // Extract the API key and Question from inputVars
  const {system, prompt, last_utterance, apiKey, model} = args.inputVars;
 
//Define the API data
  const url = 'https://api.perplexity.ai/chat/completions';
  const data = {
    model: `${model}`,
    max_tokens: 2048,
    messages: [
    {
        role: 'system',
        content: `${system}`
    },
    {
        role: 'user',
        content: ` ${prompt}. User's message: ${last_utterance}`
    }
    ],
    temperature: 0.6,
  };
 
// Wrap the API call in a try block so it can detect errors
  try {
    const response = await fetch(url, {
      method: 'POST',
      headers: {
          'content-type': 'application/json',
          'Authorization': `Bearer ${api_Key}`,
          'OpenAI-Beta': 'assistants=v1'
      },
      body: JSON.stringify(data)
    });
 
    // Check if the response status is OK (status in the range 200-299)
    if (!response.ok) {
      
    // If not OK, throw an error to be caught by the catch block
      throw new Error(`HTTP error! status: ${response.status}`);
    }
 
    // Await the response and save it to the responseBody variable
    let responseBody = await response.json;
 
    // Access the answer from the responseBody
    const answer = responseBody.choices[0].message.content;
 
 
    // Return the desired output structure
    return {
      outputVars: {
          answer: answer
      },
      next: {
          path: 'success'
      },
      trace: [
        {
          type: 'debug',
          payload: {
          message: `Information recieved: ${answer}`
                  }
        }
      ],
    };
  } catch (error) {
    // Catch block handles any errors thrown in the try block
    return {
      outputVars: {
        error: `An error occurred: ${error.message}`
      },
      next: {
        path: 'error'
      },
      trace: [
        {
          type: 'text',
          payload: {
            message: `An error occurred: ${error.message} body is ${responseBody}`
          }
        }
      ],
    };
  }
}

export default async function main(args) {
  // Extract the API key and Question from inputVars
  const {system, prompt, last_utterance, apiKey, model} = args.inputVars;
 
//Define the API data
  const url = 'https://api.perplexity.ai/chat/completions';
  const data = {
    model: `${model}`,
    max_tokens: 2048,
    messages: [
    {
        role: 'system',
        content: `${system}`
    },
    {
        role: 'user',
        content: ` ${prompt}. User's message: ${last_utterance}`
    }
    ],
    temperature: 0.6,
  };
 
// Wrap the API call in a try block so it can detect errors
  try {
    const response = await fetch(url, {
      method: 'POST',
      headers: {
          'content-type': 'application/json',
          'Authorization': `Bearer ${api_Key}`,
          'OpenAI-Beta': 'assistants=v1'
      },
      body: JSON.stringify(data)
    });
 
    // Check if the response status is OK (status in the range 200-299)
    if (!response.ok) {
      
    // If not OK, throw an error to be caught by the catch block
      throw new Error(`HTTP error! status: ${response.status}`);
    }
 
    // Await the response and save it to the responseBody variable
    let responseBody = await response.json;
 
    // Access the answer from the responseBody
    const answer = responseBody.choices[0].message.content;
 
 
    // Return the desired output structure
    return {
      outputVars: {
          answer: answer
      },
      next: {
          path: 'success'
      },
      trace: [
        {
          type: 'debug',
          payload: {
          message: `Information recieved: ${answer}`
                  }
        }
      ],
    };
  } catch (error) {
    // Catch block handles any errors thrown in the try block
    return {
      outputVars: {
        error: `An error occurred: ${error.message}`
      },
      next: {
        path: 'error'
      },
      trace: [
        {
          type: 'text',
          payload: {
            message: `An error occurred: ${error.message} body is ${responseBody}`
          }
        }
      ],
    };
  }
}

@sjst_0414 this might work? but i don't have an api key on me so i can't really test it

W. Williams (SFT)•15mo ago

@shen @sjst_0414 Re-wrote it from scratch. Works like a charm! Now supports online models and did a bit more error checking.

Perplexity_Function_...

genetic-orange•15mo ago

nice man that looks perfect

perflexity

Did you find this page helpful?