Skip to main content
GET

Query Parameters

string
required
Name of the model to inspect (e.g., “llama2”, “mistral:7b”)
boolean
default:false
Include verbose model information
string
Optional Ollama server URL (defaults to http://localhost:11434)

Response

boolean
Indicates if the request was successful
array
Model capabilities and features (e.g., [“chat”, “completion”, “embedding”])
object
Detailed model information

Notes

  • This endpoint only works for installed models (use List Models to see what’s installed)
  • The capabilities array indicates what the model can do (chat, embeddings, etc.)
  • quantization_level affects model size and performance
  • Use verbose=true for additional technical details

Common Quantization Levels

  • Q4_0: 4-bit quantization, smallest size, lower quality
  • Q5_K_M: 5-bit medium quality
  • Q8_0: 8-bit quantization, larger size, higher quality