How to show Ollama model details

Model details show what Ollama knows about a local model, including architecture, parameter size, context length, quantization, capabilities, parameters, and license. That output helps confirm the model is suitable before running a workload.

Use ollama show for terminal inspection and the api/show endpoint for automation. The output can also reveal whether a model supports tools, thinking, or embedding use cases.

Keep the full output in task evidence when model metadata changes a deployment decision. In articles and tickets, trim license or template sections that do not affect the decision.

1. List the installed model name.

   $ ollama list
   NAME           ID              SIZE
   gpt-oss:20b    17052f91a42e    13 GB

2. Show the model details.

   $ ollama show gpt-oss:20b
     Model
       architecture        gptoss
       parameters          20.9B
       context length      131072
       embedding length    2880
       quantization        MXFP4
    
     Capabilities
       completion
       tools
       thinking
   ##### snipped #####

3. Request the same model details through the API.

   $ curl -s http://localhost:11434/api/show -d '{"model":"gpt-oss:20b"}'
   {"details":{"family":"gptoss","parameter_size":"20.9B","quantization_level":"MXFP4"}}

4. Show the Modelfile when rebuilding or deriving a model.

   $ ollama show --modelfile gpt-oss:20b
   # Modelfile generated by "ollama show"
   FROM /Users/example/.ollama/models/blobs/sha256-17052f91a42e
   ##### snipped #####

5. Use the capability list to choose the right API path.

   $ printf '%s\n' 'tools thinking'
   tools thinking

   Use an embedding model for embeddings even when another model shows an embedding length.