Skip to main content

​
Synopsis

ollama show MODEL
ollama show MODEL [OPTIONS]

​
Description

The show command displays detailed information about a model, including:
  • Architecture and parameters
  • Modelfile configuration
  • System prompt and template
  • License information
  • Model capabilities
  • Quantization details

​
Arguments

​
MODEL
string
required
Name of the model to inspectExamples:
  • llama3.2
  • mistral:7b-instruct
  • myusername/custom-model

​
Options

​
--license
boolean
default:"false"
Display only the model’s license
ollama show llama3.2 --license
​
--modelfile
boolean
default:"false"
Display only the Modelfile used to create this model
ollama show llama3.2 --modelfile
​
--parameters
boolean
default:"false"
Display only the model parameters
ollama show llama3.2 --parameters
​
--template
boolean
default:"false"
Display only the prompt template
ollama show llama3.2 --template
​
--system
boolean
default:"false"
Display only the system message
ollama show llama3.2 --system
​
--verbose
boolean
default:"false"
Short: -vShow detailed model information including metadata and tensor details
ollama show llama3.2 --verbose
Only one display option (--license, --modelfile, etc.) can be used at a time.

​
Examples

​
Basic Information

Display all model information: 
ollama show llama3.2
Output: 
  Model

     architecture    llama
     parameters      3B
     context length  2048
     quantization    Q4_K_M

  Capabilities

     completion
     chat

  Parameters

     temperature     0.7
     top_p          0.9
     top_k          40

  System

     You are a helpful assistant.

​
View Modelfile

See the Modelfile used to create the model: 
ollama show llama3.2 --modelfile
Output: 
FROM llama3.2:latest

SYSTEM You are a helpful assistant.

PARAMETER temperature 0.7
PARAMETER top_p 0.9

​
View Parameters Only

Display just the parameters: 
ollama show llama3.2 --parameters
Output: 
temperature 0.7
top_p 0.9
top_k 40
repeat_penalty 1.1

​
View System Prompt

See the system message: 
ollama show llama3.2 --system
Output: 
You are a helpful assistant.

​
View Template

Display the prompt template: 
ollama show llama3.2 --template
Output: 
{{ .System }}

User: {{ .Prompt }}
Assistant:

​
View License

Display license information: 
ollama show llama3.2 --license

​
Verbose Output

Get detailed metadata and tensor information: 
ollama show llama3.2 --verbose
Output includes: 
  Model

     architecture       llama
     parameters         3B
     context length     2048
     embedding length   4096
     quantization       Q4_K_M

  Capabilities

     completion
     chat

  Metadata

     general.architecture           llama
     general.parameter_count        3000000000
     llama.context_length           2048
     llama.embedding_length         4096
     llama.attention.head_count     32
     ...

  Tensors

     token_embd.weight           F16    [4096, 32000]
     blk.0.attn_q.weight        Q4_K   [4096, 4096]
     ...

​
Output Sections

​
Model Information

Basic model details:
  • Architecture: Model architecture (llama, mistral, etc.)
  • Parameters: Number of parameters (3B, 7B, 13B, etc.)
  • Context length: Maximum context window size
  • Quantization: Quantization level applied
  • Remote model: If this is a cloud model, shows remote details

​
Capabilities

What the model can do:
  • completion - Text completion
  • chat - Conversational chat
  • embedding - Generate embeddings
  • vision - Process images (multimodal)
  • thinking - Reasoning/thinking output
  • image - Generate images

​
Parameters

Model runtime parameters:
  • temperature - Randomness in generation
  • top_p - Nucleus sampling threshold
  • top_k - Top-k sampling value
  • repeat_penalty - Penalty for repetition
  • num_ctx - Context window size
  • And more…

​
System

The system prompt that defines the model’s behavior.

​
Template

The chat template used to format prompts.

​
License

Licensing information for the model.

​
Scripting Usage

Extract specific information for scripts: 
# Get only parameters
OLLAMA_TEMP=$(ollama show llama3.2 --parameters | grep temperature | awk '{print $2}')

# Get Modelfile
ollama show mymodel --modelfile > Modelfile.backup

# Check if model has vision capability
if ollama show llama3.2 | grep -q "vision"; then
    echo "Model supports vision"
fi

​
Comparing Models

Compare two models side-by-side: 
# Compare parameters
diff <(ollama show llama3.2 --parameters) <(ollama show mistral --parameters)

# Compare Modelfiles
diff <(ollama show model1 --modelfile) <(ollama show model2 --modelfile)

​
Environment Variables

​
OLLAMA_HOST
string
default:"http://127.0.0.1:11434"
Ollama server address

​
Exit Codes

  • 0 - Success
  • 1 - Error (model not found, server not running, etc.)

​
Troubleshooting

​
Model Not Found

Error: model 'mymodel' not found
Solution: Check the model name and list available models: 
ollama list

​
Server Not Running

Error: could not connect to ollama server
Solution: Start the Ollama server: 
ollama serve

​
Use Cases

Verify Model

Confirm a model was created correctly with the right parameters

Debug Issues

Troubleshoot model behavior by inspecting configuration

Clone Models

Extract Modelfile to recreate or modify a model

Documentation

Generate documentation about your model’s configuration