OLLAMA is a free tool that allows you to run LLMS (large language models) directly on your local device. This is suitable for artificial intelligence developers or researchers, just an experiment and learning about artificial intelligence.
There is a lot to be perceived at the beginning, so the best to do is just jumping, preparing and using it. Then ask questions along the way.
Download OLLAMA
First, let’s download ollama on: For your specified operating system and then install it. When installing it, you will not see anything. It only works in the background. To interact with it, you will have to use the command line.
Interaction with ollama
Once OLLAMA is installed, open your station and type ollama --version. Here you can see that I run 0.5.11
Let’s get to know some basic orders. He writes ollama --help To show some basic orders such as:
seandotau@aseandotaus-MBP ~ % ollama --help
Large language model runner
Usage:
ollama (flags)
ollama (command)
Available Commands:
serve Start ollama
create Create a model from a Modelfile
show Show information for a model
run Run a model
stop Stop a running model
pull Pull a model from a registry
push Push a model to a registry
list List models
ps List running models
cp Copy a model
rm Remove a model
help Help about any command
Flags:
-h, --help help for ollama
-v, --version Show version informationThis driving is self -evident once it runs several times, but for the first time, let’s go through some examples.
It serves boys
This begins OLLAMA application via the command line. If you have a “graphic user interface” app, you can run this and it is not necessary to run the service order. Also, if you run other orders such as ollama listThe OLLAMA application will start automatically. All this means that you may not have to run ollama serve.
seandotau@aseandotaus-MBP ~ % ollama serve
2025/02/15 21:32:18 routes.go:1186: INFO server config env="map(HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_DEBUG:false OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST: OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:/Users/seandotau/.ollama/models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:0 OLLAMA_ORIGINS:( :* :* :* :* :* :* app://* file://* tauri://* vscode-webview://*) OLLAMA_SCHED_SPREAD:false http_proxy: https_proxy: no_proxy:)"
time=2025-02-15T21:32:18.285+11:00 level=INFO source=images.go:432 msg="total blobs: 6"
time=2025-02-15T21:32:18.285+11:00 level=INFO source=images.go:439 msg="total unused blobs removed: 0"
time=2025-02-15T21:32:18.286+11:00 level=INFO source=routes.go:1237 msg="Listening on 127.0.0.1:11434 (version 0.5.11)"
time=2025-02-15T21:32:18.335+11:00 level=INFO source=types.go:130 msg="inference compute" id=0 library=metal variant="" compute="" driver=0.0 name="" total="21.3 GiB" available="21.3 GiB"
(GIN) 2025/02/15 - 21:32:43 | 200 | 66.833µs | 127.0.0.1 | HEAD "/"
(GIN) 2025/02/15 - 21:32:43 | 200 | 1.078416ms | 127.0.0.1 | GET "/api/tags"
(GIN) 2025/02/15 - 21:32:57 | 200 | 28.167µs | 127.0.0.1 | HEAD "/"Boys list
This will list the models you downloaded. It will appear empty now, so let’s download a model.
seandotau@aseandotaus-MBP ~ % ollama list
NAME ID SIZE MODIFIEDBoil
To download a model we need to run ollama pullBut we need to know the model to be withdrawn. This is the place: He plays his role. It lists all available model options, but how do you choose?
Summary of the form
Llama: Llama is a NLP model (natural language processing) for tasks such as text generation, summary, and automatic translation. It is ideal for chatting for general purposes, artificial intelligence of conversation, and answering questions.
Mistral: Mistral, similar to Llama, treats the generation of code and data analysis on a large scale, making it ideal for developers who work on coding platforms driven by artificial intelligence
Phi-4: Microsoft’s IQ Language Model, is a 14B Teacher Modern Language Model (SLM) that excels in complex thinking in areas such as mathematics, as well as addressing traditional language.
LLAVA: LLAVA is a multimedia model capable of processing text and images. (A multimedia model is the Amnesty International system that can process and multiple types of information or “situations” – such as text, photos, sound and video – simultaneously at one time.)
Llama Code: A large language model that can use text claims to create and discuss code.
This is just a small excerpt of what is available. What you will also notice is the options of the size of the model. For example, for Llama3.2, there is a teacher size 1 billion or 3 billion. The higher the number of parameters, the greater the accuracy of the model, but the higher the size of the model. For example, Llama 3.2: 1B 1.3 GB will take it to 3.2: 3B will take 2 GB. Compare this to Llama 3.1: 140B, which will take a large area of disc 243 GB.
Until now we know the types of models and sizes of models, let’s pull Llama3.2 which is simple and lightweight. Run ollama pull llama3.2.
Note that I did not add the model teacher? If excluded, it will be hypothetical ollama pull llama3.2:latest That is currently drawing ollama pull llama3.2:3b.
seandotau@aseandotaus-MBP ~ % ollama pull llama3.2
pulling manifest
pulling dde5aa3fc5ff... 100% ▕████████████████████████████████▏ 2.0 GB
pulling 966de95ca8a6... 100% ▕████████████████████████████████▏ 1.4 KB
pulling fcc5a6bec9da... 100% ▕████████████████████████████████▏ 7.7 KB
pulling a70ff7e570d9... 100% ▕████████████████████████████████▏ 6.0 KB
pulling 56bb8bd477a5... 100% ▕████████████████████████████████▏ 96 B
pulling 34bb5ab01051... 100% ▕████████████████████████████████▏ 561 B
verifying sha256 digest
writing manifest
successShow boys
Now if you run ollama list You should see the model you have just pulled.
seandotau@aseandotaus-MBP ~ % ollama list
NAME ID SIZE MODIFIED
llama3.2:3b a80c4f17acd5 2.0 GB 28 seconds agoWhat is elegant is that you can also show information this model by operation ollama show llama3.2
seandotau@aseandotaus-MBP ~ % ollama show llama3.2
Model
architecture llama
parameters 3.2B
context length 131072
embedding length 3072
quantization Q4_K_M
Parameters
stop ""
stop ""
stop ""
License
LLAMA 3.2 COMMUNITY LICENSE AGREEMENT
Llama 3.2 Version Release Date: September 25, 2024Ollama rm
To remove a form, run it ollama rm llama3.2 For example. Advice: If you boycott a download, you will have partial files in a guide/excursion
OLLAMA PUSH, Creation, CP
These orders here may not use much unless you are creating models, copying models, or pushing them to the form record for others to use.
Ranbels
Now the moment comes. Run the form and give it a test motor. When you turn on the form, you can start interacting with it.
seandotau@aseandotaus-MBP ~ % ollama run llama3.2
>>> what is the capital of Germany?
The capital of Germany is Berlin.
>>> Send a message (/? for help)Is it cool? To go out, write /goodbye
Great additional features
Artificial intelligence is required to summarize some text
seandotau@aseandotaus-MBP ~ % ollama run llama3.2 "summarise this file in 100 words: $(cat hobbit.text )"
Here is a 100-word summary of the file:
The Hobbit, written by J.R.R. Tolkien in 1937, is a classic children's fantasy novel that has sold over 100
million copies worldwide. The story follows Bilbo Baggins, a hobbit who joins Gandalf and dwarves on a quest to
reclaim their treasure from the dragon Smaug. The book features themes of personal growth, heroism, and warfare,
drawing from Tolkien's experiences in World War I and his scholarly knowledge of Germanic philology and
mythology. Adaptations for stage, screen, radio, board games, and video games have received critical
recognition, cementing its legacy as a beloved children's fantasy novel.Ensure that the text file is in the current guide in which you are.
Artificial intelligence is required to explain the image
What you will notice here is that OLLAMA will pull the specified form first (where I had no already) and then run it.
seandotau@aseandotaus-MBP ~ % ollama run llava "What's in this image? /Users/seandotau/image.png"
pulling manifest
pulling 170370233dd5... 100% ▕██████████████████████████████████████████████████████▏ 4.1 GB
pulling 72d6f08a42f6... 100% ▕██████████████████████████████████████████████████████▏ 624 MB
pulling 43070e2d4e53... 100% ▕██████████████████████████████████████████████████████▏ 11 KB
pulling c43332387573... 100% ▕██████████████████████████████████████████████████████▏ 67 B
pulling ed11eda7790d... 100% ▕██████████████████████████████████████████████████████▏ 30 B
pulling 7c658f9561e5... 100% ▕██████████████████████████████████████████████████████▏ 564 B
verifying sha256 digest
writing manifest
success
Added image '/Users/seandotau/image.png'
The image features Pikachu, a popular Pokémon character from the franchise. It is a small electric mouse with
yellow fur, large ears, and big round eyes. Pikachu is standing upright on its hind legs, and it appears to be
looking directly at the camera.For your information: this was the picture.
Using a shaver
curl :11434/api/generate -d '{
"model": "llama3.2",
"prompt":"Why is the sky blue?"
}'Note that OLLAMA API reaches JSON responses by default, with each JSON’s response separately on a new line. This is known as JSONL Coordination (JSONL). If you want to see the full response in a more readable format, you can activate the output through jq To extract response content only:
curl :11434/api/generate
-H 'Content-Type: application/json'
-d '{"model": "llama3.2", "prompt": "Why is the sky blue?"}'
| jq -r '.response'Broadcast behavior is according to the design – it allows API to send partial responses when created, instead of waiting for a full response. If you need to disrupt the broadcast, you can add "stream": false To order your Json:
curl :11434/api/generate
-H 'Content-Type: application/json'
-d '{"model": "llama3.2", "prompt": "Why is the sky blue?", "stream": false}'
