Documentation

Antfly Inference

Local ML inference for ONNX-based models with an Ollama-compatible API — embeddings, chunking, reranking, named entity recognition, and text rewriting. Runs standalone or inside an Antfly cluster.

Getting Started

Install Antfly Inference, pull models, and serve your first requests

API Reference

Complete REST API documentation for the inference endpoints

Embedding Models

Generate text and multimodal vector embeddings

Reranking

Re-score search results for relevance

Chunking

Split text into semantically coherent segments

Kubernetes Operator

Deploy Antfly Inference on Kubernetes with autoscaling

Models

Browse available embedding, reranking, and chunking models

Downloads

Download Antfly Inference for your platform