参数化生成请求（with Options） - Ollama 官方中文文档

If you want to set custom options for the model at runtime rather than in the Modelfile, you can do so with the options parameter. This example sets every available option, but you can set any of them individually and omit the ones you do not want to override.

curl --location --request POST 'http://localhost:11434/api/generate' \ --header 'Content-Type: application/json' \ --data-raw '{ "model": "llama2-7b", "prompt": "请用中文回答以下问题", "stream": false, "options": { "num_keep": 1024, "seed": 42, "num_predict": 128, "top_k": 40, "top_p": 0.9, "temperature": 0.8, "repeat_penalty": 1.1, "stop": [ "\n", "。" ] } }'

{ "model": "llama3.2", "created_at": "2023-08-04T19:22:45.499127Z", "response": "The sky is blue because it is the color of the sky.", "done": true, "context": [1, 2, 3], "total_duration": 4935886791, "load_duration": 534986708, "prompt_eval_count": 26, "prompt_eval_duration": 107345000, "eval_count": 237, "eval_duration": 4289432000 }

请求参数

Body 参数application/json

model

string

必需

示例值:

llama2-7b

prompt

string

必需

示例值:

请用中文回答以下问题

stream

boolean

必需

示例值:

false

options

object

必需

num_keep

integer

必需

示例值:

1024

seed

integer

必需

示例值:

num_predict

integer

必需

示例值:

128

top_k

integer

必需

示例值:

top_p

number

必需

示例值:

0.9

temperature

number

必需

示例值:

0.8

repeat_penalty

number

必需

示例值:

1.1

stop

array[string]

必需

示例值:

["\n","。"]

示例

返回响应

🟢200Success

application/json

Body

model

string

必需

created_at

string

必需

response

string

必需

done

boolean

必需

context

array[integer]

必需

total_duration

integer

必需

load_duration

integer

必需

prompt_eval_count

integer

必需

prompt_eval_duration

integer

必需

eval_count

integer

必需

eval_duration

integer

必需