Admin: Created page with "= Ollama Modelfile Cheat Sheet = == Basic Structure == A Modelfile is a recipe for building a model.

 FROM   # Optional settings PARAMETER    SYSTEM   TEMPLATE   ADAPTER   LICENSE   MESSAGE

== FROM (Required) == Base model you build on.

 FROM llama3 FROM mistral FROM ./local-model.gguf

== PARAMETER == Controls runtime behavior. === Common Parameters ===

 PARAMETER temperatur..."

2026-05-01T18:04:54Z

Created page with "= Ollama Modelfile Cheat Sheet = == Basic Structure == A Modelfile is a recipe for building a model. <pre> FROM <base_model> # Optional settings PARAMETER <key> <value> SYSTEM <text> TEMPLATE <text> ADAPTER <path> LICENSE <text> MESSAGE <role> <text> </pre> == FROM (Required) == Base model you build on. <pre> FROM llama3 FROM mistral FROM ./local-model.gguf </pre> == PARAMETER == Controls runtime behavior. === Common Parameters === <pre> PARAMETER temperatur..."

New page

= Ollama Modelfile Cheat Sheet =

== Basic Structure ==
A Modelfile is a recipe for building a model.

<pre>
FROM <base_model>

# Optional settings
PARAMETER <key> <value>

SYSTEM <text>

TEMPLATE <text>

ADAPTER <path>

LICENSE <text>

MESSAGE <role> <text>
</pre>

== FROM (Required) ==
Base model you build on.

<pre>
FROM llama3
FROM mistral
FROM ./local-model.gguf
</pre>

== PARAMETER ==
Controls runtime behavior.

=== Common Parameters ===

<pre>
PARAMETER temperature 0.7
PARAMETER top_p 0.9
PARAMETER top_k 40
PARAMETER num_ctx 4096
PARAMETER repeat_penalty 1.1
PARAMETER seed 42
PARAMETER num_predict 128
</pre>

=== Description ===

* '''temperature''' – randomness (0 = deterministic, 1 = highly random)
* '''top_p''' – nucleus sampling probability cutoff
* '''top_k''' – limits token selection pool
* '''num_ctx''' – context window size
* '''repeat_penalty''' – reduces repetition
* '''num_predict''' – max tokens to generate

== SYSTEM ==
Defines assistant behavior.

<pre>
SYSTEM You are a helpful assistant that explains things simply.
</pre>

== TEMPLATE ==
Controls prompt formatting.

<pre>
TEMPLATE """
{{ if .System }}<|system|>
{{ .System }}
{{ end }}

<|user|>
{{ .Prompt }}

<|assistant|>
"""
</pre>

== MESSAGE ==
Few-shot examples.

<pre>
MESSAGE user What is 2+2?
MESSAGE assistant 4
</pre>

== ADAPTER ==
Attach LoRA adapters.

<pre>
ADAPTER ./my-lora
</pre>

== LICENSE ==
Attach a license.

<pre>
LICENSE MIT
</pre>

== Full Example ==

<pre>
FROM llama3

PARAMETER temperature 0.7
PARAMETER top_p 0.9
PARAMETER num_ctx 4096

SYSTEM You are a precise and concise coding assistant.

TEMPLATE """
{{ if .System }}<|system|>
{{ .System }}
{{ end }}

<|user|>
{{ .Prompt }}

<|assistant|>
"""

MESSAGE user Write a Python function to add two numbers.
MESSAGE assistant def add(a, b): return a + b
</pre>

== Build and Run ==

<pre>
ollama create mymodel -f Modelfile
ollama run mymodel
</pre>

== Common Pitfalls ==

* Missing FROM → build fails
* Too large num_ctx → high memory usage
* Incorrect TEMPLATE → broken responses
* High temperature → hallucinations
* Low temperature → overly rigid output

== Mental Model ==

* FROM = base model
* SYSTEM = behavior
* TEMPLATE = prompt structure
* PARAMETER = tuning
* MESSAGE = examples
* ADAPTER = extensions

Ollama Modelfile Sample - Revision history