> ## Documentation Index
> Fetch the complete documentation index at: https://docs.pawa-ai.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Models

> Pawa Models are a new family of African-built AI small language models (SLM) developed by Sartify Company Limited to open new wave of innovations in Africa. Learn about the capabilities, and how to choose the right one for your use case.

<div className="relative rounded-2xl overflow-hidden bg-neutral-900 shadow-lg min-h-[400px] flex">
  <div className="flex-1 p-8 text-white flex flex-col justify-between">
    <div>
      <h2 className="text-3xl font-bold mb-4">Introducing Pawa Blaze</h2>

      <p className="text-neutral-200 text-[0.9rem] mb-6 leading-relaxed">
        We're thrilled to introduce Pawa Blaze, our latest advanced small language model with reasoning, agentic, multimodal, knowledge-heavy tasks and optimized
        for multilingual understanding and long-context applications.
      </p>

      {/* Key Information Grid */}

      <div className="grid grid-cols-2 gap-6 mb-6">
        <div>
          <h3 className="text-lg font-medium text-neutral-400 mb-1">Modalities</h3>
          <p className="text-sm font-semibold">Text, Image → Text</p>
        </div>

        <div>
          <h3 className="text-lg font-medium text-neutral-400 mb-1">Context window</h3>
          <p className="text-sm font-semibold">1M</p>
        </div>
      </div>

      {/* Features List */}

      <div className="mb-6">
        <h3 className="text-lg font-medium text-neutral-400 mb-3">Features</h3>

        <div className="space-y-2">
          <div className="flex items-center space-x-2">
            <span className="text-lg">
              <svg xmlns="http://www.w3.org/2000/svg" width="24" height="24" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round" class="lucide lucide-dollar-sign h-4 text-[hsl(var(--neutral98))]">
                <line x1="12" x2="12" y1="2" y2="22" />

                <path d="M17 5H9.5a3.5 3.5 0 0 0 0 7h5a3.5 3.5 0 0 1 0 7H6" />
              </svg>
            </span>

            <span className="text-neutral-200 text-sm">Low cost</span>
          </div>

          <div className="flex items-center space-x-2">
            <span className="text-lg">
              <svg xmlns="http://www.w3.org/2000/svg" width="24" height="24" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round" class="lucide lucide-earth h-4 text-[hsl(var(--neutral98))]">
                <circle cx="12" cy="12" r="10" />

                <path d="M2 12h20" />

                <path d="M12 2a15.3 15.3 0 0 1 0 20" />

                <path d="M12 2a15.3 15.3 0 0 0 0 20" />
              </svg>
            </span>

            <span className="text-neutral-200 text-sm">Multilingual</span>
          </div>

          <div className="flex items-center space-x-2">
            <span className="text-lg">
              <svg xmlns="http://www.w3.org/2000/svg" width="24" height="24" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round" class="lucide lucide-lightbulb h-4 text-[hsl(var(--neutral98))]">
                <path d="M15 14c.2-1 .7-1.7 1.5-2.5 1-.9 1.5-2.2 1.5-3.5A6 6 0 0 0 6 8c0 1 .2 2.2 1.5 3.5.7.7 1.3 1.5 1.5 2.5" />

                <path d="M9 18h6" />

                <path d="M10 22h4" />
              </svg>
            </span>

            <span className="text-neutral-200 text-sm">Reasoning</span>
          </div>

          <div className="flex items-center space-x-2">
            <span className="text-lg">
              <svg xmlns="http://www.w3.org/2000/svg" width="24" height="24" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round" class="lucide lucide-square-function h-4 text-[hsl(var(--neutral98))]">
                <rect width="18" height="18" x="3" y="3" rx="2" ry="2" />

                <path d="M9 17c2 0 2.8-1 2.8-2.8V10c0-2 1-3.3 3.2-3" />

                <path d="M9 11.2h5.7" />
              </svg>
            </span>

            <span className="text-neutral-200 text-sm">Tools calling</span>
          </div>
        </div>
      </div>
    </div>

    <button onClick={() => window.open("https://builder.pawa-ai.com/dashboard?page=sandbox", "_blank", "noopener,noreferrer")} className="bg-[#FFA200] text-black px-3 py-1 rounded-lg font-medium transition-colors w-fit inline-block hover:bg-[#e69500]">
      Test model
    </button>
  </div>

  <div
    className="flex-1 relative"
    style={{
  backgroundImage: 'url(https://mintcdn.com/sartifycoltd/QqnQOhdEkmPzvUEE/images/blaze_noun_004_0366.jpg?fit=max&auto=format&n=QqnQOhdEkmPzvUEE&q=85&s=efecf68af093e9ccc319178c2b8b155e)',
  backgroundSize: 'cover',
  backgroundPosition: 'center',
  backgroundRepeat: 'no-repeat'
}}
  >
    {/* Blur overlay on the left side where it meets the text content */}

    <div className="absolute inset-0 bg-gradient-to-r from-neutral-900/40 via-transparent to-transparent" />

    <div
      className="absolute left-0 top-0 bottom-0 w-full"
      style={{
    backdropFilter: 'blur(8px)',
    background: 'linear-gradient(to right, rgba(23, 23, 23, 0.7), rgba(23, 23, 23, 0.3))'
  }}
    />
  </div>
</div>

## Models Categories

The Pawa Models family is built to power applications ranging from chatbots and assistants to document analysis, search, reasoning and agentic systems. The following are the categories of our models

* **Chat Models**: For chat, agentic, tools calling, reasoning, multilingual & multimodal. Eg.*Pawa Ember*, *Pawa Blaze*.
* **Text to Speech**: For converting text to hearable audio. Eg. *Pawa Text To Speech*,
* **Speech to Text**: For converting audio to readable text. Eg. *Pawa Speech To Text*.
* **Embeddings**: For converting audio to readable text. Eg. *Pawa Embeddings*
* **Documents Parsing**: For converting audio to readable text. *Pawa Parsing*,

### Chat Models

Chat models are designed to handle conversational AI, reasoning, multilingual support, agentic and multimodal inputs. They are ideal for building chatbots, virtual assistants, and agentic systems capable of tool calling and complex problem solving.

<div className="overflow-x-auto">
  <table className="min-w-full border-collapse">
    <thead>
      <tr className="bg-neutral-800 text-white text-left">
        <th className="px-4 py-2">Model Name</th>
        <th className="px-4 py-2">Alias Name</th>
        <th className="px-4 py-2">Description</th>
        <th className="px-4 py-2">Input Type</th>
        <th className="px-4 py-2">Output Type</th>
        <th className="px-4 py-2">Context Length</th>
        <th className="px-4 py-2">Status</th>
      </tr>
    </thead>

    <tbody>
      <tr className="bg-neutral-900 even:bg-neutral-800 text-neutral-200">
        <td className="px-4 py-2">pawa-v1-ember-20240924</td>
        <td className="px-4 py-2">Pawa Ember</td>
        <td className="px-4 py-2">Funny & great for everyday tasks</td>
        <td className="px-4 py-2">text</td>
        <td className="px-4 py-2">text</td>
        <td className="px-4 py-2">100K</td>
        <td className="px-4 py-2">ACTIVE</td>
      </tr>

      <tr className="bg-neutral-900 even:bg-neutral-800 text-neutral-200">
        <td className="px-4 py-2">pawa-v1-blaze-20250318</td>
        <td className="px-4 py-2">Pawa Blaze</td>
        <td className="px-4 py-2">Fast, visual, tools & most intelligent</td>
        <td className="px-4 py-2">text, image</td>
        <td className="px-4 py-2">text</td>
        <td className="px-4 py-2">1M</td>
        <td className="px-4 py-2">ACTIVE</td>
      </tr>
    </tbody>
  </table>
</div>

### Voice Models

Chat models are designed to handle conversational AI, reasoning, multilingual support, agentic and multimodal inputs. They are ideal for building chatbots, virtual assistants, and agentic systems capable of tool calling and complex problem solving.

<div className="overflow-x-auto">
  <table className="min-w-full border-collapse">
    <thead>
      <tr className="bg-neutral-800 text-white text-left">
        <th className="px-4 py-2">Model Name</th>
        <th className="px-4 py-2">Alias Name</th>
        <th className="px-4 py-2">Description</th>
        <th className="px-4 py-2">Input Type</th>
        <th className="px-4 py-2">Output Type</th>
        <th className="px-4 py-2">Status</th>
      </tr>
    </thead>

    <tbody>
      <tr className="bg-neutral-900 even:bg-neutral-800 text-neutral-200">
        <td className="px-4 py-2">pawa-stt-v1-20240701</td>
        <td className="px-4 py-2">Pawa Speech To Text</td>
        <td className="px-4 py-2">Converts spoken audio into written text for transcription or analysis.</td>
        <td className="px-4 py-2">audio</td>
        <td className="px-4 py-2">text</td>
        <td className="px-4 py-2">ACTIVE</td>
      </tr>

      <tr className="bg-neutral-900 even:bg-neutral-800 text-neutral-200">
        <td className="px-4 py-2">pawa-tts-v1-20250704</td>
        <td className="px-4 py-2">Pawa Text To Speech</td>
        <td className="px-4 py-2">Generates high-quality audio from written text for narration or voice assistants.</td>
        <td className="px-4 py-2">text</td>
        <td className="px-4 py-2">audio</td>
        <td className="px-4 py-2">ACTIVE</td>
      </tr>
    </tbody>
  </table>
</div>

### Embedding Models

Chat models are designed to handle conversational AI, reasoning, multilingual support, agentic and multimodal inputs. They are ideal for building chatbots, virtual assistants, and agentic systems capable of tool calling and complex problem solving.

<div className="overflow-x-auto">
  <table className="min-w-full border-collapse">
    <thead>
      <tr className="bg-neutral-800 text-white text-left">
        <th className="px-4 py-2">Model Name</th>
        <th className="px-4 py-2">Alias Name</th>
        <th className="px-4 py-2">Description</th>
        <th className="px-4 py-2">Input Type</th>
        <th className="px-4 py-2">Output Type</th>
        <th className="px-4 py-2">Status</th>
      </tr>
    </thead>

    <tbody>
      <tr className="bg-neutral-900 even:bg-neutral-800 text-neutral-200">
        <td className="px-4 py-2">pawa-embeddings-v1-20241001</td>
        <td className="px-4 py-2">Pawa Embeddings</td>
        <td className="px-4 py-2">Transforms text into numerical vectors for search, similarity, or AI reasoning tasks.</td>
        <td className="px-4 py-2">text</td>
        <td className="px-4 py-2">embeddings</td>
        <td className="px-4 py-2">ACTIVE</td>
      </tr>
    </tbody>
  </table>
</div>

### Parsing Models

Chat models are designed to handle conversational AI, reasoning, multilingual support, agentic and multimodal inputs. They are ideal for building chatbots, virtual assistants, and agentic systems capable of tool calling and complex problem solving.

<div className="overflow-x-auto">
  <table className="min-w-full border-collapse">
    <thead>
      <tr className="bg-neutral-800 text-white text-left">
        <th className="px-4 py-2">Model Name</th>
        <th className="px-4 py-2">Alias Name</th>
        <th className="px-4 py-2">Description</th>
        <th className="px-4 py-2">Input Type</th>
        <th className="px-4 py-2">Output Type</th>
        <th className="px-4 py-2">Status</th>
      </tr>
    </thead>

    <tbody>
      <tr className="bg-neutral-900 even:bg-neutral-800 text-neutral-200">
        <td className="px-4 py-2">pawa-v1-parser-20250809</td>
        <td className="px-4 py-2">Pawa Parsing</td>
        <td className="px-4 py-2">Extracts text and structured data from various document formats for analysis or AI processing.</td>
        <td className="px-4 py-2">documents</td>
        <td className="px-4 py-2">text</td>
        <td className="px-4 py-2">ACTIVE</td>
      </tr>
    </tbody>
  </table>
</div>

## Additional Information Regarding Models

#### 1.Use `web_search_tool` for auto realtime updated data

Pawa chat models have no knowledge of current events or data beyond what was present in their training data.\
To incorporate realtime data with your request, please use builtin web\_search\_tool, or pass any realtime data using custom tools or RAG as context in your system prompt.

#### 2. For chat models, roles should alternate. Don't send two same roles mutiple times, use contents instead, to add other informations on that role.

Don't send two same roles mutiple times, use contents instead, to add other informations on that role for your conversation context.

### 3. For Image input models

* Maximum image size: 25MiB
* Maximum number of images: No limit
* Supported image file types: `jpg/jpeg` or `png`
* Any image/text input order is accepted (e.g. text prompt can precede image prompt)

### 3. For Embedding models

* Maximum document size: 15MiB
* Maximum number of documents: 15, upgrade to get many more.
* Supported file types: `jpg/jpeg`, `png`, \`\`

### 4. For Voice models

* Maximum document size: 15MiB
* Maximum number of documents: 15, upgrade to get many more.
* Supported file types: `jpg/jpeg`, `png`, \`\`

### 5. For Parsing models

* Maximum document size: 15MiB
* Maximum number of documents: 15, upgrade to get many more.
* Supported file types: `jpg/jpeg`, `png`, \`\`

***

## Next Steps

* [API Reference](/api-reference/introduction)
* [What can i build ?](/learn-more/what-can-you-build)