Quickstart

This guide describes steps to get you to start developing or testing the Nebula models quickly.

Account Access

Make sure you have access to Nebula. You can request access using sign up form. Once your request is approved, you will receive instructions and access details. You'll need an API Key to make calls to Nebula.

You can use the Model API playground to start testing the model quickly without the need to write any code. The playground is a great way to get started by testing your transcripts and prompts.

Make your first API call

Make sure you have cURL or Python installed on your machine to use the following examples. If you're using python please install the requests library to make the API calls.

Once you have an API key, and cURL or Python is setup, you can make an API call. The following example shows a simple example to send a message to the Chat model.

export NEBULA_API_KEY="YOUR_API_KEY"
curl --location "https://api-nebula.symbl.ai/v1/model/chat" \
--header "ApiKey: $NEBULA_API_KEY" \
--header "Content-Type: application/json" \
--data '{
		"max_new_tokens": 1024,
    "messages": [
        {
            "role": "human",
            "text": "Hello, I am Mark. Give me few tips on how to handle objections in sales call."
        }
    ]
}'

# Copy this code in a nebula-test.py file and run using -
# python nebula-test.py

import requests
import json

NEBULA_API_KEY="YOUR_API_KEY"  # Replace with your API key

url = "https://api-nebula.symbl.ai/v1/model/chat"

payload = json.dumps({
  "max_new_tokens": 1024,
  "messages": [
    {
      "role": "human",
      "text": "Hello, I am Mark. Give me few tips on how to handle objections better in sales calls."
    }
  ]
})
headers = {
  'ApiKey': NEBULA_API_KEY,
  'Content-Type': 'application/json'
}

response = requests.request("POST", url, headers=headers, data=payload)

print(response.json())

In this example, we're asking the model to provide open-ended tips to handle objections in a sales call.

💡Try including a transcript of a call as part of your message for the model to refer to, and try changing message to perform other tasks.

Try including a transcript of a sales call as part of your message for the model to refer to.

You can also try the Instruct model with your transcript with a command to model.

export NEBULA_API_KEY="YOUR_API_KEY"
curl --location "https://api-nebula.symbl.ai/v1/model/generate" \
--header "ApiKey: $NEBULA_API_KEY" \
--header "Content-Type: application/json" \
--data '{
    "max_new_tokens": 1024,
    "prompt": {
    	"instruction": "Generate a summary of the conversation emphasizing on Customer'\''s pain points.",
      "conversation": {
      	"text": "Customer: Hi Mark, I'\''m really glad to speak with you today. I'\''m facing some challenges with our current cloud provider, and I wanted to share my concerns with you.\nMark: Of course, I'\''m here to help. Please feel free to explain the issues you'\''re experiencing, and we'\''ll see how we can address them.\nCustomer: Thank you. Well, we'\''ve been using this provider for a while now, but we'\''ve noticed that their services have been slow and unreliable lately. It'\''s affecting our business operations, and we'\''re losing productivity due to frequent downtime and performance issues.\nMark: I understand. What specific issues are you encountering when using their services?\nCustomer: There are a few things. First, their server response times are really slow, and it'\''s affecting our website'\''s loading speed. Our customers have been complaining about the site being unresponsive, which is impacting our user experience negatively.\nMark: That'\''s definitely a concern. What else?\nCustomer: Additionally, we'\''ve been experiencing data loss and inconsistencies. We'\''ve lost critical data a couple of times, and it'\''s affecting our data integrity and team morale. It'\''s becoming difficult to trust them with our sensitive information.\nMark: I see. Data loss is a serious issue. Are you considering any specific cloud providers in mind?\nCustomer: Yes, we'\''ve been researching a few options, and we'\''re interested in exploring DataFly. We'\''ve heard good things about your services and would like to know if you can help us with a smooth migration process.\nMark: Absolutely, we'\''d be happy to help. Our team can ensure a seamless migration and provide you with reliable, fast, and secure cloud services. Let'\''s discuss the details and find the best solution for your business."
      }
    }
}'

# Copy this code in a nebula-test.py file and run using -
# python nebula-test.py

import requests
import json

NEBULA_API_KEY="YOUR_API_KEY"  # Replace with your API key

url = "https://api-nebula.symbl.ai/v1/model/generate"

payload = json.dumps({
  "max_new_tokens": 1024,
  "prompt": {
    "instruction": "Generate a summary of the conversation emphasizing on Customer's pain points.",
    "conversation": {
      "text": "Customer: Hi Mark, I'm really glad to speak with you today. I'm facing some challenges with our current cloud provider, and I wanted to share my concerns with you.\nMark: Of course, I'm here to help. Please feel free to explain the issues you're experiencing, and we'll see how we can address them.\nCustomer: Thank you. Well, we've been using this provider for a while now, but we've noticed that their services have been slow and unreliable lately. It's affecting our business operations, and we're losing productivity due to frequent downtime and performance issues.\nMark: I understand. What specific issues are you encountering when using their services?\nCustomer: There are a few things. First, their server response times are really slow, and it's affecting our website's loading speed. Our customers have been complaining about the site being unresponsive, which is impacting our user experience negatively.\nMark: That's definitely a concern. What else?\nCustomer: Additionally, we've been experiencing data loss and inconsistencies. We've lost critical data a couple of times, and it's affecting our data integrity and team morale. It's becoming difficult to trust them with our sensitive information.\nMark: I see. Data loss is a serious issue. Are you considering any specific cloud providers in mind?\nCustomer: Yes, we've been researching a few options, and we're interested in exploring DataFly. We've heard good things about your services and would like to know if you can help us with a smooth migration process.\nMark: Absolutely, we'd be happy to help. Our team can ensure a seamless migration and provide you with reliable, fast, and secure cloud services. Let's discuss the details and find the best solution for your business."
    }
  }
})
headers = {
  'ApiKey': NEBULA_API_KEY,
  'Content-Type': 'application/json'
}

response = requests.request("POST", url, headers=headers, data=payload)

print(response.json())