# Pawa AI Docs

- [Introduction](https://docs.pawa-ai.com/index.md): Solving Africa’s Biggest Untapped AI Potential—**Localization**
- [Quickstart](https://docs.pawa-ai.com/get-started/quickstart.md): Get started with Pawa AI and integrate powerful AI capabilities in minutes
- [Glossary](https://docs.pawa-ai.com/get-started/glossary.md): A reference of key terms and concepts used throughout Pawa AI or AI industry in general
- [Changelog](https://docs.pawa-ai.com/get-started/changelog.md): Stay up-to-date with the latest features, releases, and improvements in Pawa AI.
- [Models](https://docs.pawa-ai.com/learn-more/models.md): 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.
- [What to build ?](https://docs.pawa-ai.com/learn-more/what-can-you-build.md): Explore practical use cases you can build powered by Pawa AI with ease and low cost in Africa. Learn how our APIs can help you build intelligent applications, automate workflows, and enhance decision-making.
- [Features](https://docs.pawa-ai.com/learn-more/feature.md): Pawa AI offers different features, whether you’re a **developer**, a **founder**, or an **enterprise team**, Pawa AI helps you build intelligent applications out of the box.
- [Languages Supported](https://docs.pawa-ai.com/learn-more/languages.md): Pawa AI is built for inclusive intelligence, supporting local and global languages so anyone can access AI regardless of where they are or what language they speak.
- [Pricing](https://docs.pawa-ai.com/learn-more/pricing.md): Simple, transparent pricing with **Subscription** plans or **Pay as you go** credit tiers. No unexpected costs due to token-based billing.
- [Payment Type Switch](https://docs.pawa-ai.com/learn-more/payment-type-switch.md): Learn about Pawa AI's two account payment types—**Subscription** and **Pay as you go**—and how to switch between them anytime without losing data.
- [Chat Requests](https://docs.pawa-ai.com/capabilities/chat-and-text-generation.md): With Pawa AI APIs you can use our models that has been fine-tuned to follow instructions and respond to natural language prompts through chat using prompt.
- [In Memory Chat](https://docs.pawa-ai.com/capabilities/in-memory-chat.md): With Pawa AI APIs you can make your app or system remember previous conversations by processing the context of your request history.
- [RAG](https://docs.pawa-ai.com/capabilities/rag.md): Pawa AI APIs make it easy to build Retrieval-Augmented Generation (RAG) systems out of the box, without unnecessary complexity or technical jargon and at low cost.
- [Agentic RAG](https://docs.pawa-ai.com/capabilities/agentic-rag.md): Pawa AI APIs enable you to build advanced Agentic Retrieval-Augmented Generation (RAG) systems out of the box—simplifying complexity, reducing cost, and delivering smarter results.
- [Audio & Transcription](https://docs.pawa-ai.com/capabilities/audio-transcription.md): Pawa AI APIs provide seamless audio `input and output capabilities` — `Speech-to-Text (STT)` and `Text-to-Speech (TTS)`. These enable models to chat, speak, and understand audio directly, supporting use cases like voice assistants, transcription, and audio-enhanced knowledge bases.
- [Web Browsing](https://docs.pawa-ai.com/capabilities/web-browsing.md): Pawa AI APIs enable access and retrieval of real-time information from the web. With browsing capabilities, your assistant can go beyond its training data to fetch the latest news, check live events, validate facts, and provide up-to-date answers.
- [Vision](https://docs.pawa-ai.com/capabilities/vision.md): Pawa AI models’ vision capabilities enable analyzing images and providing insights based on visual content, in addition to text. This multimodal approach opens up new possibilities for applications that require both textual and visual understanding.
- [Tools Calling](https://docs.pawa-ai.com/capabilities/tools-calling.md): Give the model actions using function call. Define function call with structured arguments, to make Pawa AI models communicate with external systems and access data outside their training data.
- [Reasoning](https://docs.pawa-ai.com/capabilities/reasoning.md): Pawa AI help getting higher‑quality answers on complex tasks by enabling structured internal reasoning.
- [Embeddings](https://docs.pawa-ai.com/capabilities/embeddings.md): Pawa AI can create the higher‑quality vector representations of text that capture the semantic meaning of paragraphs through their position in a high-dimensional vector space.
- [Semantic Retrieval](https://docs.pawa-ai.com/capabilities/semantic-retrieval.md): Pawa AI helps find relevant documents and information using natural language queries instead of exact keyword matches.
- [Documents Parsing](https://docs.pawa-ai.com/capabilities/documents-parsing.md): Pawa AI offers enterprise-level document processing, combining cutting-edge OCR technology with advanced structured data extraction from any files such as images, audio, pdfs, docx, html, pptsx, xlsx, etc...Experience faster processing speeds, unparalleled accuracy, and cost-effective solutions, all…
- [Knowledge Base](https://docs.pawa-ai.com/capabilities/knowledge-base.md): Pawa AI Knowledge Base capability offers a managed, scalable vectors document store with semantic indexing and retrieval. Upload files, search by meaning, and power your innovative RAG in minutes.
- [Streaming Responses](https://docs.pawa-ai.com/capabilities/streaming.md): Pawa AI enables continuous generation of partial results from the model within moments, even if the full generation takes longer.
- [Structured Output](https://docs.pawa-ai.com/capabilities/structured.md): Pawa AI structured output capabilities gives a way to define, request, and get JSON structured outputs directly from models for more reliable and machine-readable responses.
- [Finetuning (beta)](https://docs.pawa-ai.com/capabilities/finetuning.md): Customize Pawa AI models to your domain, brand, and data. Train chat and voice models with your own datasets for more accurate, personalized outputs.
- [Introduction](https://docs.pawa-ai.com/agents/introduction.md): An overview of agents in Pawa AI — autonomous, task-oriented systems that can reason, act, and interact with users and tools.
- [Create Agents](https://docs.pawa-ai.com/agents/create-agents.md): Pawa AI lets you create autonomous, task-oriented agents that can reason, act, and interact with users and external tools. These agents can perform complex workflows, automate tasks, and provide intelligent assistance across different domains.
- [Agents Conversations](https://docs.pawa-ai.com/agents/agents-conversations.md): Learn how to initiate and manage conversations with Pawa AI agents, including context handling, multi-agent routing, and configurable parameters for generating intelligent responses.
- [Agents Handoff](https://docs.pawa-ai.com/agents/agents-handoff.md): Learn how Pawa AI handles multi-agent orchestration through agent handoff, where a RouterAgent routes queries to the most suitable specialized agent for execution.
- [Agents Best Practices](https://docs.pawa-ai.com/agents/agent-best-practices.md): Recommended strategies and design patterns for building reliable, efficient, and effective agents with Pawa AI.
- [Best Practices](https://docs.pawa-ai.com/going-production/deployment.md): Comprehensive guidelines for securely, efficiently, and responsibly using the Pawa AI API in production environments.
- [Limits](https://docs.pawa-ai.com/api-reference/limits.md): Pawa AI have put several limitations to ensure optimal performance and a positive experience on APIs for all users.
- [Temperature](https://docs.pawa-ai.com/guides/temperature.md): A comprehensive guide to understanding and using the temperature parameter in language model requests, including practical recommendations, examples, and advanced tips.
- [Top-p (Nucleus Sampling)](https://docs.pawa-ai.com/guides/top-p.md): A comprehensive guide to understanding and using top-p (nucleus sampling) in LM requests, including practical examples, best practices, and advanced tips.
- [Frequency and Presence Penalties](https://docs.pawa-ai.com/guides/frequency-and-presence-penalty.md): An in-depth guide on understanding and using frequency and presence penalties in LM requests to control repetition, improve diversity, and optimize output quality.
- [Max Tokens](https://docs.pawa-ai.com/guides/max-tokens.md): Understanding and controlling the maximum number of tokens a model can generate.
- [Observability](https://docs.pawa-ai.com/guides/obervability.md): Understanding observability in AI systems for monitoring, debugging, and improving performance.
- [Prompt Engineering](https://docs.pawa-ai.com/guides/prompt-engineering.md): Techniques and best practices for designing effective prompts for AI models.
- [Community](https://docs.pawa-ai.com/resources/community.md): Join the Pawa AI community on Hugging Face.
- [Blueprints Codes](https://docs.pawa-ai.com/resources/blueprints-codes.md): Explore Pawa AI blueprint code examples on GitHub.
- [Feedback](https://docs.pawa-ai.com/resources/feedback.md): Share feedback with the Pawa AI team on WhatsApp.
- [Terms of Service](https://docs.pawa-ai.com/resources/terms-of-service.md): Read the Pawa AI terms and conditions.
- [Privacy Policy](https://docs.pawa-ai.com/resources/privacy-policy.md): Read the Pawa AI privacy policy.
- [Introduction](https://docs.pawa-ai.com/api-reference/introduction.md): The Pawa AI APIs are a core component of our mission to empower  organizations to build intelligent, AI-powered solutions with ease.  They are designed to support teams of any size, providing robust integration  capabilities that allow for customization and enable organizations to derive  maximum va…
- [Authentication](https://docs.pawa-ai.com/api-reference/authentication.md): Pawa AI APIs requires API Keys as authentication to ensure secure access and usage. This section details the methods and best practices for authenticating your requests.
- [Limits](https://docs.pawa-ai.com/api-reference/limits.md): Pawa AI have put several limitations to ensure optimal performance and a positive experience on APIs for all users.
- [Introduction](https://docs.pawa-ai.com/api-reference/endpoint/models/introduction.md): Pawa AI model includes powerful small language models in language, voice, audio, multimodal, embeddings, parsing etc... such as Pawa Ember and Pawa Blaze that supports swahili, english and other African languages. Our ecosystem is designed to support innovation at every level, while reduce cost of A…
- [Get All Models](https://docs.pawa-ai.com/api-reference/endpoint/models/list.md): This endpoint used to get all the models registered in the Pawa AI Platform with pagination, search, and filtering support.
- [Get A Single Model](https://docs.pawa-ai.com/api-reference/endpoint/models/single.md): This endpoint used to get the model by id in the Pawa AI Platform.
- [Introduction](https://docs.pawa-ai.com/api-reference/endpoint/chat/introduction.md): The Pawa AI chat endpoint provides a unified interface for conversation, reasoning, multimodal input, retrieval-augmented generation (RAG), agentic workflows, tools, and structured outputs — all powered by our Pawa model family.
- [Send Request](https://docs.pawa-ai.com/api-reference/endpoint/chat/request.md): This endpoint is used to receive chat request from user through API and return the answer.
- [Introduction](https://docs.pawa-ai.com/api-reference/endpoint/voice/introduction.md): Pawa AI provides powerful voice capabilities with Text-to-Speech (TTS) and Speech-to-Text (STT), enabling natural, real-time, multilingual voice interaction for your applications.
- [Convert Text to speech](https://docs.pawa-ai.com/api-reference/endpoint/voice/text-to-speech.md): This endpoint converts text into streaming audio. It does streaming automatically so consider sending request as events streams, Server-Sent Events (SSE) or any other streaming method.
- [Convert Speech to text](https://docs.pawa-ai.com/api-reference/endpoint/voice/speech-to-text.md): This endpoint converts a list files into text for each of one
- [Introduction](https://docs.pawa-ai.com/api-reference/endpoint/agents/introduction.md): The Pawa API helps you create, manage, and interact with autonomous AI agents. Agents can reason, use tools, perform RAG, and collaborate with chat or external workflows.
- [Create Agents](https://docs.pawa-ai.com/api-reference/endpoint/agents/create.md): This endpoint used to create the new custom agent to the Pawa AI Platform.
- [Agents Conversation](https://docs.pawa-ai.com/api-reference/endpoint/agents/conversation.md): This endpoint is used to receive agent chat request from user through API and return the answer.
- [Agents Handoff](https://docs.pawa-ai.com/api-reference/endpoint/agents/handoff.md): This endpoint is used to receive agent chat request from user through API and return the answer.
- [Edit Agent](https://docs.pawa-ai.com/api-reference/endpoint/agents/update.md): This endpoint used to update the custom agent in the Pawa AI Platform.
- [Delete Agent](https://docs.pawa-ai.com/api-reference/endpoint/agents/delete.md): This endpoint used to delete the custom agent in the Pawa AI Platform.
- [Get All Agents](https://docs.pawa-ai.com/api-reference/endpoint/agents/view.md): This endpoint used to retrieve all custom agents created in the Pawa AI Platform.
- [Get Single Agent](https://docs.pawa-ai.com/api-reference/endpoint/agents/view-single.md): This endpoint used to get the custom agent by id in the Pawa AI Platform.
- [Introduction](https://docs.pawa-ai.com/api-reference/endpoint/vectors/introduction.md): The Pawa AI Vectors API lets you create a high-quality embeddings for languages that can be used for retrieval-augmented generation (RAG), semantic search, personalization, and knowledge grounding.
- [Create Embeddings](https://docs.pawa-ai.com/api-reference/endpoint/vectors/embeddings.md): This endpoint used to create new custom embeddings for the given input text using the specified model.
- [Introduction](https://docs.pawa-ai.com/api-reference/endpoint/documents/introduction.md): The Pawa AI Documents API lets you parse or extract any contents from files and documents. This include getting contents from audio, images, text, pdf, docx, excel, ppt.
- [Parse Documents](https://docs.pawa-ai.com/api-reference/endpoint/documents/parse.md): This endpoint used to parse the documents and extract text from them.
- [Introduction](https://docs.pawa-ai.com/api-reference/endpoint/storage/introduction.md): The Pawa AI Knowledge Base API lets you create, manage, and query documents for agents, RAG, and semantic search applications.
- [Create Knowledge Base](https://docs.pawa-ai.com/api-reference/endpoint/storage/knowledge-base/create.md): This endpoint used to create the new knowledge base in the Pawa AI Platform.
- [Get All Knowledge Bases](https://docs.pawa-ai.com/api-reference/endpoint/storage/knowledge-base/get-all.md): This endpoint used to get all the knowledge bases registered in the Pawa AI Platform with pagination, search, and filtering support.
- [Semantic Retrieval](https://docs.pawa-ai.com/api-reference/endpoint/storage/knowledge-base/semantic-retrival.md): This endpoint used to retrieve the relevant chunks of the knowledge base based on the user question in the Pawa AI Platform.
- [Get a Knowledge Base](https://docs.pawa-ai.com/api-reference/endpoint/storage/knowledge-base/get-single.md): This endpoint used to get the knowledge base by id in the Pawa AI Platform.
- [Delete a Knowledge Base](https://docs.pawa-ai.com/api-reference/endpoint/storage/knowledge-base/delete.md): This endpoint used to delete the knowledge base by id.
- [Edit a Knowledge Base](https://docs.pawa-ai.com/api-reference/endpoint/storage/knowledge-base/edit.md): This endpoint used to update the knowledge base in the Pawa AI Platform.
- [Get Files of KB](https://docs.pawa-ai.com/api-reference/endpoint/storage/knowledge-base/get-files.md): This endpoint used to list of files used by knowledge base id.
- [Introduction](https://docs.pawa-ai.com/api-reference/endpoint/transcribe/introduction.md): The Pawa AI Transcribe API lets you use an all-in-one transcription capabilities with unlimited audio and video processing. Convert audio and video to accurate text in seconds.
- [Create Workspace](https://docs.pawa-ai.com/api-reference/endpoint/transcribe/create-workspace.md): This endpoint is used to create a new workspace for the authenticated user.
- [Get All Workspaces](https://docs.pawa-ai.com/api-reference/endpoint/transcribe/get-workspaces.md): This endpoint is used to retrieve all workspaces created by the authenticated user.
- [Get A Single Workspace](https://docs.pawa-ai.com/api-reference/endpoint/transcribe/get-workspace.md): This endpoint is used to retrieve a specific workspace by its ID.
- [Update Workspace](https://docs.pawa-ai.com/api-reference/endpoint/transcribe/update-workspace.md): This endpoint is used to update the name of an existing workspace.
- [Delete Workspace](https://docs.pawa-ai.com/api-reference/endpoint/transcribe/delete-workspace.md): This endpoint is used to delete a workspace and all its associated files.
- [Upload Transcriptions](https://docs.pawa-ai.com/api-reference/endpoint/transcribe/upload-transcriptions.md): This endpoint is used to upload transcriptions to a specific workspace.
- [Get Workspace Transcriptions](https://docs.pawa-ai.com/api-reference/endpoint/transcribe/get-transcriptions.md): This endpoint is used to retrieve all transcriptions belonging to a specific workspace with pagination support.
- [Update Transcription](https://docs.pawa-ai.com/api-reference/endpoint/transcribe/update-transcription.md): This endpoint is used to update the transcript content of a specific transcription within a workspace.
- [Delete Transcription](https://docs.pawa-ai.com/api-reference/endpoint/transcribe/delete-transcription.md): This endpoint is used to remove a specific transcription from a workspace.
- [Introduction](https://docs.pawa-ai.com/python-sdk/introduction.md): Official Python SDK for the Pawa AI API — install, authenticate, and call every endpoint from models to transcribe.
- [Async, retries & errors](https://docs.pawa-ai.com/python-sdk/async-retries-errors.md): Async client, retry configuration, and error handling for the Pawa AI Python SDK.
- [Models](https://docs.pawa-ai.com/python-sdk/models.md): List and retrieve Pawa AI models with the Python SDK.
- [Chat](https://docs.pawa-ai.com/python-sdk/chat.md): Chat completions with the Python SDK — basic, streaming, tools, structured output, vision, reasoning, and RAG.
- [Voice](https://docs.pawa-ai.com/python-sdk/voice.md): Text-to-speech and speech-to-text with the Pawa AI Python SDK.
- [Vectors](https://docs.pawa-ai.com/python-sdk/vectors.md): Create embeddings with the Pawa AI Python SDK.
- [Documents](https://docs.pawa-ai.com/python-sdk/documents.md): Parse documents with the Pawa AI Python SDK.
- [Agents](https://docs.pawa-ai.com/python-sdk/agents.md): Create, manage, and chat with agents using the Pawa AI Python SDK.
- [Storage](https://docs.pawa-ai.com/python-sdk/storage.md): Manage knowledge bases and run semantic retrieval with the Python SDK.
- [Transcribe](https://docs.pawa-ai.com/python-sdk/transcribe.md): Manage transcription workspaces and uploads with the Pawa AI Python SDK.

## OpenAPI Specs

- [openapi](/api-reference/openapi.json)

## Optional

- [SandBox](https://builder.pawa-ai.com/dashboard?page=sandbox)
- [Get Your API Key](https://builder.pawa-ai.com/dashboard?page=keys)
- [Examples](https://github.com/Sartify/pawa-ai-blueprints-chatbots)
- [Blogs](https://pawa-ai.com/blogs)


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.