{"id":"ollama","name":"Ollama","tagline":"Run open models on your own machine with one command.","description":"The easiest way to serve Llama, Mistral and friends locally; a US company maintains it.","website":"https://ollama.com","source":"https://github.com/ollama/ollama","license":"MIT","self_hostable":true,"vendor":{"id":"ollama","name":"Ollama Inc.","country":"US","parent":null,"kind":"company","url":"https://ollama.com","european":false},"needs":["ai-models"],"replaces":["openai-api"],"offerings":[{"mode":"self-host","price":{"basis":"free"}}],"resources":{"ram_mb":4096,"cpu":2,"disk_gb":20,"basis":"a 7B model on CPU; a GPU makes it pleasant"},"ash":"ollama","kind":"software","provenance":{"source":"curated","verified":"2026-09-04"},"open_source":true,"data_classes":["intellectual-property"],"modes":["sh"],"verdicts":{"sh":{"verdict":"green","rule":1,"reason":"You run it yourself: no data leaves your own infrastructure."}},"upstream":{"source":"awesome-selfhosted-data","license":["MIT"],"stars":180079,"updated":"2026-09-03","platforms":["Docker","Python"],"tags":["Generative Artificial Intelligence (GenAI)"],"archived":false,"description":"Get up and running with Llama 3.3, DeepSeek-R1, Phi-4, Gemma 3, and other large language models."},"needs_expanded":[{"id":"ai-models","name":"AI models","data_classes":["intellectual-property"]}],"page":"/tools/ollama"}