Create a model
Body
Name for the model to create
Existing model to create from
Prompt template to use for the model
Name of the renderer for the model
Name of the parser for the model
Source file names mapped to their SHA-256 digests. Split GGUF models must include each shard under its original split filename.
Draft source file names mapped to their SHA-256 digests
License string or list of licenses for the model
System prompt to embed in the model
Key-value parameters for the model
Message history to use for the model
Quantization level to apply during import (e.g. nvfp4)
Quantization level to apply to draft weights during import
Minimum Ollama version required by the model
Capabilities to add without removing inherited or inferred capabilities (e.g. decision for compatible decision models)
Stream status updates
Response
Stream of create status updates
Current status message