Ollama
- Category
- AI and LLMs
- Docs outline
- 15 sections
Official docs outline
docs.ollama.com- Quickstart
Running a model locally with ollama run and the basic local server model.
- CLI reference
Commands such as pull, run, ps, list, show, create, and serve.
- Modelfile reference
FROM, PARAMETER, TEMPLATE, SYSTEM, and ADAPTER instructions for building custom models.
- REST API
The local HTTP API at port 11434, endpoints, streaming, and conventions.
- Streaming and API usage
Streaming responses and how request and response fields work.
- Structured outputs
Constraining output to a JSON schema.
- Tool calling
Letting models call functions and feeding results back.
- Embeddings and vision
Generating embeddings for retrieval and multimodal image input.
- Thinking models
Reasoning models and controlling thinking output.
- OpenAI compatibility
Using Ollama through OpenAI-style client libraries and endpoints.
- Context length
Setting num_ctx, memory impact, and default context behavior.
- Hardware and GPU support
GPU backends, supported accelerators, and CPU fallback.
- Importing models
Importing GGUF and Safetensors models and quantization.
- FAQ and operations
Keep-alive, concurrency, environment variables, network exposure, and storage locations.
- Ollama cloud
Running larger models on hosted infrastructure through the same API.
Each section links to the official page. Titles and summaries by Extra Credit; the docs belong to their authors.
