Ollama

Ollama

Modo debug

/set verbose

Prompt: Hello

>>> Hello

Resposta:

Hello! How are you today? Is there something I can help you with, or would you like to chat?

total duration:       678.164416ms
load duration:        18.576248ms
prompt eval count:    11 token(s)
prompt eval duration: 131.281ms
prompt eval rate:     83.79 tokens/s
eval count:           24 token(s)
eval duration:        486.965ms
eval rate:            49.28 tokens/s

Configurações

Prompt:

>>> /set system "Responsa em portugues, o mais preciso possível, e no máximo em duas linhas"

Resposta:

Set system message.

Sobre o modelo

Prompt:

>>> /show

Response:

Available Commands:
 /show info         Show details for this model
 /show license      Show model license
 /show modelfile    Show Modelfile for this model
 /show parameters   Show parameters for this model
 /show system       Show system message
 /show template     Show prompt template

Prompt 2:

>>> /show info

Sair do modulo

Prompt:

>>> /clear

Resposta:

Cleared session context

Ollama Comandos

Listas os modelos existentes

ollama list

Removento modelos existentes

ollama rm gemma2:27b

Criando um modelo

ollama create brapci

Modelo do arquivo de criação (modelo.txt):

FROM llama3.3:latest

PARAMETER temperature 0.1
PARAMETER num_ctx 1024

SYSTEM Responsa em no máximo 2 linhas em portugues. Respostas curtas e precisas.

Criando o modelo Prompt:

ollama create brapci -f ./model.txt

Resposta:

transferring model data
using existing layer sha256:8eeb52dfb3bb9aefdf9d1ef24b3bdbcfbe82238798c4b918278320b6fcef18fe
using existing layer sha256:948af2743fc78a328dcb3b0f5a31b3d75f415840fdb699e8b1235978392ecf85
using existing layer sha256:0ba8f0e314b4264dfd19df045cde9d4c394a52474bf92ed6a3de22a4ca31a177
creating new layer sha256:55dcf8076ae23604428d8b4e3da7b4cf1f7901d56cd5c372b35019e89fc55329
creating new layer sha256:dd8a35e7060436c1dfed155f80c14ea5cb012a4a3c0c460bf4087ceffdb62d4d
creating new layer sha256:dd7674e7d8fe16dad131bde90382cc861eaa698bf78a259215512f68d55be608
writing manifest
success

Ollama API

Generator

POST /api/generate
[https://github.com/ollama/ollama/blob/main/docs/api.md#generate-a-completion]

Chat

 POST /api/chat 
[https://github.com/ollama/ollama/blob/main/docs/api.md#generate-a-chat-completion]