AI Inference API Providers

Fully managed AWS service providing foundation models from Anthropic, Meta, Mistral, Cohere, and more. OpenAI-compatible API with enterprise-grade security and compliance.

Official Claude API. Direct access to Claude Opus, Sonnet, and Haiku models.

Ultra-fast AI inference on custom Wafer-Scale Engine chips. Up to 3000 tok/s output speed, 20x faster than GPU-based providers.

Serverless inference for open-source LLMs and generative models. Pay-per-token with fast cold starts.

Fast serverless inference for open-source models with per-token pricing, fine-tuning, and on-demand deployments.

Official Gemini API via Google AI Studio and Vertex AI. Direct access to Gemini, Imagen, and Gemma models.

Fastest LLM inference powered by custom LPU chips. OpenAI-compatible API with sub-second latency.

Affordable AI API aggregator offering 259+ models across chat, image, video, and music at discounted prices.

Official Mistral API. Direct access to Mistral Large, Small, and Ministral models. EU data residency available.

AI API aggregator with 315+ model endpoints across text, image, video, and audio at competitive prices.

Budget AI inference platform with broad model catalog across LLM, image, video, and audio. Very competitive per-token pricing.

Official OpenAI API. Direct access to GPT, DALL-E, Whisper, and embedding models.

Unified API for 300+ LLMs from OpenAI, Anthropic, Google, Meta, and more. Routes to the best provider automatically.

Run and deploy machine learning models with a cloud API. Pay-per-use with serverless GPU infrastructure.

Fast and affordable AI inference platform. 2.3x faster speeds and 32% lower latency than major cloud platforms. Supports LLM, image, video, and audio models.

Serverless and dedicated inference for open-source LLMs, image, video, and audio models. GPU clusters available.

Fast inference platform for generative media — image, video, audio, and 3D models with serverless GPU infrastructure.

Unified API for 400+ AI models across text, image, video, and audio. OpenAI-compatible with serverless inference.

Alibaba Cloud Model Studio offers the Qwen model family including LLMs, image generation (Qwen Image, Wan), video generation (Wan), embeddings, and TTS.

Full-modal AI inference platform with 300+ models. Smart routing to cheapest servers with transparent pay-as-you-go pricing.

Creator of the Flux model family. Offers image generation and editing via a credit-based API. Models range from real-time (Klein) to highest quality (Max).

ByteDance's international AI platform offering Seedream image generation and Seedance video generation models via the ModelArk API.

Edge AI inference across 200+ cities worldwide. Serverless, pay-per-use with OpenAI-compatible API.

Creator of the Command and Embed model families. Enterprise-focused NLP platform with LLMs, embeddings, and reranking. OpenAI-compatible API.

Creator of DeepSeek V3 and R1 models. Offers extremely competitive pricing on high-quality LLMs with thinking/reasoning modes. OpenAI-compatible API.

Creator of industry-leading text-to-speech, speech-to-text, music generation, and sound effects models. Credit-based pricing with free tier.

Fast serverless and dedicated AI inference. Korean provider with competitive pricing on open-source models and prompt caching support.

Open-access AI cloud with serverless inference and dedicated hosting. Zero data retention, privacy-first design. Unified per-token pricing.

Creator of the Kling video and image generation models by Kuaishou. Offers text-to-video, image-to-video, and image generation via API.

Creator of Ray 2 video generation and Photon image generation models. Pixel-based pricing with upscaling and audio options.

Creator of Hailuo video generation, MiniMax M2.5/M2.7 LLMs, Music 2.5, and Speech models. Full multimodal API platform with Anthropic-compatible endpoints.

Creator of the Kimi model family. Known for long-context LLMs with strong code and agent capabilities. OpenAI-compatible API at platform.moonshot.ai.

European AI inference on Token Factory. Two flavors: fast (low latency) and base (cost-efficient). Batch at 50% off.

Distributed AI inference network with serverless, dedicated, and batch endpoints. No rate limits, no contracts, up to 30x cheaper than legacy cloud.

Creator of the Arrow vector SVG generation models. Offers text-to-SVG and image-to-SVG (vectorize) via REST API with credit-based pricing. Raised $8.3M seed led by a16z.

Creator of the Recraft image generation models. Offers raster and SVG vector image generation with per-image API pricing.

Creator of Gen-4.5 and Aleph video generation models. Credit-based API at $0.01/credit. Also offers image generation and audio models.

AI inference on custom SN50 chips. OpenAI-compatible API with fast output speeds. GPT OSS 120B at 600+ tok/s.

Privacy-focused AI inference with no logging. Supports open-source and proprietary models including Claude, GPT, Grok, and open-weight LLMs.

Creator of the Voyage embedding and reranking models. Specializes in high-quality text embeddings with domain-specific models for code, law, and finance.

Fast AI inference platform specializing in image and video generation with serverless GPU infrastructure.

Creator of the GLM model family. Offers LLMs, vision models, image generation (CogView), and video generation (CogVideoX) via an OpenAI-compatible API at z.ai.

Creator of the Grok model family. Offers LLMs with reasoning modes, image generation, and video generation. Batch API available at 50% discount.