Quickstart

Follow these steps to start using Pawa AI models in your applications.

Step 1: Create a Pawa AI Account

Sign up for a Pawa AI account Register.
Verify your email and log in as a Builder

Step 2: Generate an API Key

Navigate to the API Keys page in your Builders Dashboard.
Click Create new key.
Copy the key and save it securely. We recommend storing it as an environment variable:

export PAWA_AI_API_KEY="your_api_key_here"

Step 3: Make Your First API Call

Chat
Text to Speech
Speech to Text
Embeddings
Document Parsing

With your API key set, you can make your first chat request:

curl https://api.pawa-ai.com/v1/chat/request \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $PAWA_AI_API_KEY" \
  -d '{
    "model": "pawa-v1-ember-20240924",
    "messages": [
      {
        "role": "system",
        "content": "You are a TanzaBot, a helpful AI assistant for developers in Africa."
      },
      {
        "role": "user",
        "content": "Hello, How can I use AI in my app?"
      }
    ],
    "stream": false
  }'

You should receive a JSON response with Pawa’s chat completion.

Convert text to speech:

curl https://api.pawa-ai.com/v1/voice/text-to-speech \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $PAWA_AI_API_KEY" \
  -d '{
    "model": "pawa-tts-v1-20250704",
    "text": "Hello, this is Pawa AI speaking!",
    "voice": "liora"
  }'

You’ll receive an audio file in the response. We technically do auto streaming of speech(text to speech), so if need that please send request as as streaming using server sent events(SSE).

Convert speech to text:

curl https://api.pawa-ai.com/v1/voice/speech-to-text \
  -H "Authorization: Bearer $PAWA_AI_API_KEY" \
  -F "file=@audio_file.wav" \
  -F "model=pawa-stt-v1-20240701"

The response will contain the transcribed text.

Generate embeddings for NLP tasks:

curl https://api.pawa-ai.com/v1/vectors/embedding \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $PAWA_AI_API_KEY" \
  -d '{
    "model": "pawa-embeddings-v1-20241001",
    "input": ["Embed this sentence.", "And this one too."],
    "lang":"multi"
  }'

The response will include numerical vectors for each input text. The list of vector representations of each sentence.

Parse documents to extract text:

curl https://api.pawa-ai.com/v1/documents/parse \
  -H "Authorization: Bearer $PAWA_AI_API_KEY" \
  -F "documents=@document.pdf" \
  -F "model=pawa-v1-parser-20250809"

The response will contain the extracted textual data.

Next Steps

Now that your account is ready and you’ve made your first requests, explore other Pawa AI’s capabilities:

Tools calling

Enable your AI to call external tools and APIs for enhanced functionality.

Reasoning

Build AI systems that can think through complex problems step by step.

Agents

Create autonomous AI agents that can perform multi-step tasks independently.

Multimodal

Process text, images, and other media types in a single AI conversation.

Need help? Check the API Reference or visit our Sandbox to test the models.

Getting Started

Learn More

Capabilities

Agents

Going Production

Guides

Resources

Step 1: Create a Pawa AI Account

Step 2: Generate an API Key

Step 3: Make Your First API Call

Next Steps

Tools calling

Reasoning

Agents

Multimodal

Getting Started

Learn More

Capabilities

Agents

Going Production

Guides

Resources

​Step 1: Create a Pawa AI Account

​Step 2: Generate an API Key

​Step 3: Make Your First API Call

​Next Steps

Tools calling

Reasoning

Agents

Multimodal

Step 1: Create a Pawa AI Account

Step 2: Generate an API Key

Step 3: Make Your First API Call

Next Steps