frameworksUpdated: 2026-11-01
Ollama
ওলামা (Ollama)
Ollama is an open-source local LLM runtime and model package manager that wraps llama.cpp into a zero-config daemon exposing an OpenAI-compatible /v1 API on port 11434.
Overview
Ollama packages GGUF model weights, chat templates, and system parameters into Docker-style Modelfile bundles. By exposing http://localhost:11434/v1/chat/completions and /v1/embeddings, it acts as a zero-cost local drop-in replacement for OpenAI and Anthropic APIs inside n8n, LangGraph, AutoGen, and Cursor.