AI APIs
API documentation, SDKs, and usage guides for mainstream AI services
The Complete OpenAI API Guide
A comprehensive, step-by-step guide to the OpenAI API ecosystem, covering GPT, DALL·E, and Whisper models in depth. It walks you through account setup, authentication, making your first API call, prompt design patterns, and cost optimisation strategies—ideal for developers new to the OpenAI platform who want a structured learning path.
Anthropic Claude API
The API gateway to Anthropic's Claude model family, widely praised for strong reasoning, coding ability, and reliable handling of long documents. Key capabilities include an extended context window, tool use and function calling for agentic workflows, structured output, and prompt caching. It is built with safety and reliability as core design principles, making it a strong choice for developers who need careful, trustworthy AI behaviour.
Replicate API
A cloud platform that lets you run thousands of open-source AI models through a simple API call, covering image generation, video, audio, and text. You do not need to manage GPUs or any infrastructure — just send a request and Replicate handles the rest. It follows a pay-per-use pricing model, making it ideal for developers who want to prototype quickly without hardware commitments.
Anthropic Claude API Documentation
The official documentation for the Anthropic Claude API, providing comprehensive guides on authentication, request formatting, and model capabilities. It covers Claude's strengths in reasoning, coding, and long-document analysis, along with pricing details, rate limits, and best practices for production use. Aimed at developers integrating Claude into applications that require careful, reliable AI behaviour.
Google Gemini API Documentation
The official documentation for Google's Gemini API, covering text, image, audio, and video understanding in a unified multimodal interface. It includes guides on authentication, function calling, context caching, and code execution, with a generous free tier for experimentation. Aimed at developers building applications that need to process multiple content types within a single request.
Replicate API Usage Guide
A step-by-step tutorial on using the Replicate API to run thousands of open-source AI models for image, video, audio, and text tasks. It covers account setup, making your first prediction, understanding pricing, and optimising for cold starts. Ideal for developers who want to experiment with diverse models without committing to GPU infrastructure.
DeepSeek API Usage Guide
DeepSeek provides high-performance chat and reasoning models at remarkably low cost, with an OpenAI-compatible API that makes migration trivial. This guide walks through setup, key endpoints, and best practices for getting the most out of DeepSeek's strong coding and math capabilities.
Mistral AI API
Mistral AI offers a high-performance LLM API featuring both open-source and commercial models with strong multilingual capabilities. The API supports function calling, JSON mode, and long-context processing, providing a flexible and cost-effective alternative for developers building chat, coding, and reasoning applications.
Groq High-Speed Inference API
Groq delivers ultra-fast LLM inference through its custom LPU hardware, offering API access to popular open-source models at remarkably low latency. It is ideal for real-time applications like chatbots and code assistants where speed matters, with pricing based on token usage.
Cohere API
Cohere provides an enterprise-grade API for text generation, semantic search, embedding, and reranking, optimized for building high-quality search and RAG pipelines. Its models excel at understanding business context, and the platform offers straightforward integration with popular vector databases and orchestration frameworks.
OpenRouter API
OpenRouter provides a unified API that routes requests to hundreds of large language models from multiple providers through a single OpenAI-compatible endpoint. It simplifies model switching, offers transparent pricing, and handles failover automatically — ideal for developers who want flexibility without managing multiple API keys.
Together AI API
Together AI is a cloud platform and API for running, fine-tuning, and deploying open-source AI models with fast, cost-effective inference. It offers a wide catalog of community models, serverless endpoints, and dedicated capacity options, making it easy to build production applications without managing GPU infrastructure.
Perplexity Sonar API
Perplexity Sonar is an API that delivers grounded, real-time answers powered by web-wide search with inline citations. It is designed for developers building AI search experiences, chatbots, and research tools that need up-to-date, verifiable information rather than relying solely on a model's training data.
Qwen (Tongyi Qianwen) API
Qwen is Alibaba's large language model API, offering strong multilingual and Chinese-language capabilities across chat, coding, and vision models. Many Qwen models are also released with open weights, giving developers the choice between hosted API convenience and self-deployed control at competitive prices.
ERNIE (Wenxin) API
ERNIE is Baidu's large model API with particularly strong Chinese-language understanding, multimodal support, and deep enterprise integrations within the Baidu ecosystem. It suits businesses targeting the Chinese market that need reliable language processing, knowledge-enhanced reasoning, and compliance with local regulations.
Zhipu GLM API
Zhipu AI's GLM series is a competitive Chinese large model family accessible via API, with strong reasoning, coding, and agent capabilities. Spun out of Tsinghua University research, GLM models offer both hosted API access and open-weight releases, making them popular for Chinese-language applications and agent development.
xAI Grok API
Grok is xAI's large model API featuring real-time knowledge through X platform integration, large context windows, and strong reasoning performance. It offers developers an alternative to established providers, with distinctive strengths in current-events awareness and a personality-driven conversational style.
Kimi (Moonshot AI)
Kimi is Moonshot AI's assistant and API, known for very long context handling and strong reasoning performance — a leading choice in the Chinese model ecosystem. It excels at processing lengthy documents and complex analysis, with competitive API pricing for developers building Chinese-language applications.
Azure OpenAI Service
Azure OpenAI Service provides OpenAI models on Microsoft Azure with enterprise-grade security, private networking, regional deployment options, and compliance controls. It suits organizations that need GPT-level capabilities within existing Azure governance, data residency requirements, and Microsoft ecosystem integrations.
Amazon Bedrock
Amazon Bedrock is a managed AWS service offering foundation models from multiple providers — including Anthropic, Meta, and Amazon — behind one unified API. It adds agents, guardrails, and knowledge bases for building production AI applications with AWS security, monitoring, and pay-as-you-go pricing.
Google Vertex AI
Vertex AI is Google Cloud's unified AI platform, bringing together Gemini and partner models, training and tuning pipelines, agent building tools, and MLOps in one place. It targets enterprises that want to develop, deploy, and govern AI applications entirely within the Google Cloud ecosystem.
Tencent Hunyuan
Hunyuan is Tencent's foundation model family spanning chat, image, video, and 3D generation, with several open-weight releases alongside APIs on Tencent Cloud. Its video and 3D models are particularly notable in the open-source community, making it a versatile option for multimodal Chinese-market applications.
MiniMax
MiniMax is a Chinese frontier AI lab offering text, speech, music, and video models through a unified API, known especially for long-context language models and high-quality voice synthesis. It provides developers a versatile multimodal toolkit at competitive prices, popular for voice agents and creative applications.
iFlytek Spark
Spark is iFlytek's foundation model built on decades of speech technology leadership, offering industry-leading Chinese speech recognition and synthesis alongside general language capabilities. It is widely deployed in education, healthcare, and enterprise scenarios across China, with APIs on the iFlytek open platform.
HyperCLOVA X
HyperCLOVA X is Naver's Korean-first foundation model powering the CLOVA service family, with APIs for chat, embeddings, and Korean-optimized applications. Trained with deep Korean language and cultural understanding, it outperforms global models on Korean tasks — the natural choice for services targeting the Korean market.
AI21 Jamba
Jamba is AI21 Labs' Mamba-Transformer hybrid model family, combining state-space efficiency with attention quality to handle very long contexts at lower cost. Available as open weights and through an enterprise API, it suits document-heavy workloads like contract analysis and long-form summarization.