AI・機械学習

Whisper (STT) and TTS

Command models with RAG and tool use

Multimodal AI — text, image, video understanding

GPT models — chat, vision, function calling

Ultra-fast LLM inference — Llama, Mixtral, Gemma

AI search with real-time web grounding

GPT models with enterprise security on Azure

Grok models via OpenAI-compatible API

Run 200+ open models — Llama, Mixtral, DeepSeek, SDXL

Fast inference for open-source models

Ultra-fast inference — 2,100 tok/s on Llama 3.3

Fast inference on custom RDU chips

Text embeddings via Voyager models

Text embeddings for search and classification

Run embedding models via Inference API

Gemini models — multimodal AI

Generate and edit video with Gen-3 Alpha

Generate images with excellent text rendering

DALL-E image generation and editing

Run 300,000+ open-source models via API

Call Claude, Llama, Mistral, and more from a single API

Open-source LLMs — download or use via hosting providers

Claude models via Messages API — text, vision, tool use

Generate songs from text prompts

Sentiment, entities, keywords, categories extraction

Enterprise LLM for business writing

Empathetic personal AI assistant

Run open-source models — SDXL, Llama, Whisper

Real-time TTS — 40ms latency, streaming

Real-time and pre-recorded transcription — 50+ languages

Realistic text-to-speech with voice cloning

Ultra-realistic TTS with voice cloning — 142 languages

Real-time and async transcription with speaker labels

Clone voices from audio samples