跳过至正文
Ollama 提供了与部分 OpenAI API 的兼容性，以帮助将现有应用程序连接到 Ollama。

​
用法

​
简单的 /v1/chat/completions 示例

basic.py
from openai import OpenAI

client = OpenAI(
    base_url='https://:11434/v1/',
    api_key='ollama',  # required but ignored
)

chat_completion = client.chat.completions.create(
    messages=[
        {
            'role': 'user',
            'content': 'Say this is a test',
        }
    ],
    model='gpt-oss:20b',
)
print(chat_completion.choices[0].message.content)

​
简单的 /v1/responses 示例

responses.py
from openai import OpenAI

client = OpenAI(
    base_url='https://:11434/v1/',
    api_key='ollama',  # required but ignored
)

responses_result = client.responses.create(
  model='qwen3:8b',
  input='Write a short poem about the color blue',
)
print(responses_result.output_text)

​
带有视觉功能的 /v1/chat/completions 示例

vision.py
from openai import OpenAI

client = OpenAI(
    base_url='https://:11434/v1/',
    api_key='ollama',  # required but ignored
)

response = client.chat.completions.create(
    model='qwen3-vl:8b',
    messages=[
        {
            'role': 'user',
            'content': [
                {'type': 'text', 'text': "What's in this image?"},
                {
                    'type': 'image_url',
                    'image_url': 'data:image/png;base64,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',
                },
            ],
        }
    ],
    max_tokens=300,
)
print(response.choices[0].message.content)

​
端点

​
/v1/chat/completions

​
支持的功能

  • 对话补全 (Chat completions)
  • 流式输出
  • JSON 模式
  • 可复现的输出
  • 视觉
  • 工具 (Tools)
  • 推理/思考控制（适用于思考模型）
  • Logprobs (对数概率)

​
支持的请求字段

  • model
  • messages
    • 文本 content
    • 图像 content
      • Base64 编码图像
      • 图像 URL
    • content 部分数组
  • frequency_penalty (频率惩罚)
  • presence_penalty (存在惩罚)
  • response_format (响应格式)
  • seed (种子)
  • stop (停止词)
  • stream
  • stream_options (流选项)
    • include_usage (包含使用情况)
  • temperature
  • top_p
  • max_tokens (最大 Token 数)
  • tools
  • reasoning_effort (推理力度："high", "medium", "low", "none")
  • reasoning (推理)
    • effort (力度："high", "medium", "low", "none")
  • tool_choice (工具选择)
  • logit_bias
  • user (用户标识)
  • n

​
/v1/completions

​
支持的功能

  • 文本补全 (Completions)
  • 流式输出
  • JSON 模式
  • 可复现的输出
  • Logprobs (对数概率)

​
支持的请求字段

  • model
  • prompt (提示词)
  • frequency_penalty (频率惩罚)
  • presence_penalty (存在惩罚)
  • seed (种子)
  • stop (停止词)
  • stream
  • stream_options (流选项)
    • include_usage (包含使用情况)
  • temperature
  • top_p
  • max_tokens (最大 Token 数)
  • suffix (后缀)
  • best_of
  • echo
  • logit_bias
  • user (用户标识)
  • n

​
备注

  • prompt 目前仅接受字符串

​
/v1/models

​
备注

  • created 对应于模型最后修改的时间
  • owned_by 对应于 ollama 用户名，默认为 "library"

​
/v1/models/{model}

​
备注

  • created 对应于模型最后修改的时间
  • owned_by 对应于 ollama 用户名，默认为 "library"

​
/v1/embeddings

​
支持的请求字段

  • model
  • input (输入)
    • string
    • 字符串数组
    • Token 数组
    • Token 数组的数组
  • encoding format (编码格式)
  • dimensions (维度)
  • user (用户标识)

​
/v1/images/generations (实验性)

注意：此端点处于实验阶段，可能会在未来版本中更改或移除。
使用图像生成模型生成图像。
images.py
from openai import OpenAI

client = OpenAI(
    base_url='https://:11434/v1/',
    api_key='ollama',  # required but ignored
)

response = client.images.generate(
    model='x/z-image-turbo',
    prompt='A cute robot learning to paint',
    size='1024x1024',
    response_format='b64_json',
)
print(response.data[0].b64_json[:50] + '...')

​
支持的请求字段

  • model
  • prompt (提示词)
  • size (例如 “1024x1024”)
  • response_format (仅支持 b64_json)
  • n
  • quality (质量)
  • style (风格)
  • user (用户标识)

​
/v1/responses

注意：在 Ollama v0.13.3 中添加
Ollama 支持 OpenAI Responses API。仅支持非状态化版本（即不支持 previous_response_id 或 conversation）。

​
支持的功能

  • 流式输出
  • 工具（函数调用）
  • 推理摘要（适用于思考模型）
  • 有状态请求

​
支持的请求字段

  • model
  • input (输入)
  • instructions (指令)
  • tools
  • stream
  • temperature
  • top_p
  • max_output_tokens (最大输出 Token 数)
  • previous_response_id (不支持有状态 v1/responses)
  • conversation (不支持有状态 v1/responses)
  • truncation (截断)

​
模型

在使用模型之前，请先通过 ollama pull 本地拉取它 
ollama pull llama3.2

​
默认模型名称

对于依赖 gpt-3.5-turbo 等默认 OpenAI 模型名称的工具，请使用 ollama cp 将现有模型名称复制到临时名称 
ollama cp llama3.2 gpt-3.5-turbo
随后，可以在 model 字段中指定这个新模型名称 
curl https://:11434/v1/chat/completions \
    -H "Content-Type: application/json" \
    -d '{
        "model": "gpt-3.5-turbo",
        "messages": [
            {
                "role": "user",
                "content": "Hello!"
            }
        ]
    }'

​
设置上下文窗口大小

OpenAI API 无法设置模型的上下文窗口大小。如果您需要更改上下文大小，请创建一个类似于以下内容的 Modelfile 
FROM <some model>
PARAMETER num_ctx <context size>
使用 ollama create mymodel 命令创建具有更新后上下文窗口大小的新模型。使用更新后的模型名称调用 API 
curl https://:11434/v1/chat/completions \
    -H "Content-Type: application/json" \
    -d '{
        "model": "mymodel",
        "messages": [
            {
                "role": "user",
                "content": "Hello!"
            }
        ]
    }'