Ollama Server ExplorationDirect API DiscoveryWCAG AAA Accessible
Ollama Model Explorer
Connect to any local or remote Ollama server to discover, audit, and explore downloaded model weights, parameter families, quantization tiers, and execution commands.
Server Connection
Specify the HTTP endpoint exposed by the Ollama background daemon.
Common Hosts:
Ollama CORS Tip: Launch with
OLLAMA_ORIGINS="*" ollama serveDiscovered Weights0 of 0 Available
Model Catalog & Metadata
Ready for Server Exploration
Provide the host address above and click Fetch Models to query the Ollama engine, or toggle the sample preview dataset.
Developer Command Cheatsheet
Quick terminal commands to pull, run, and administer models on your target machine:
Pull Model
ollama pull llama3:8bRun Interactive REPL
ollama run deepseek-r1:8bList In-Memory VRAM
ollama ps