AI Inference API Providers
Compare AI inference API providers side by side — pricing, supported models, features, and free tiers. Whether you need the cheapest LLM API, the fastest image generation endpoint, or a provider with OpenAI-compatible routing, find the right fit below.
Covering 5+ providers including OpenRouter, Together AI, Fireworks AI, fal.ai, Replicate, DeepInfra, Groq, and more. Filter by type, category, or browse the full directory.
KIE AI
Affordable AI API aggregator offering 120+ models across chat, image, video, and music at discounted prices, often well below official API rates.
Muapi
AI API aggregator with 315+ model endpoints across text, image, video, and audio at competitive prices.
OpenRouter
Unified API for 300+ LLMs from OpenAI, Anthropic, Google, Meta, and more. Routes to the best provider automatically.
fal.ai
Fast inference platform for generative media — image, video, audio, and 3D models with serverless GPU infrastructure.
Token Harbor
AI model gateway: one OpenAI- and Anthropic-compatible API across frontier chat models from Anthropic, OpenAI, Google, xAI, Moonshot, Alibaba, Z.ai, DeepSeek, Xiaomi and Muse. Transparent per-token pricing (vendor price + per-model markup, 0% default), semantic/exact response caching, th-orchestra agentic model routing, no silent model swaps. Free monthly allowance on selected models; optional Agent/Office/Frontier passes add included usage and 5-15% pay-as-you-go discounts.