【AI-Ollama入门】Ollama完整API

notes / AI notes
字数统计: 7.4k   |   阅读时长≈ 37 分钟

【❤】Ollama完整的API

规范

模型名称

  • 模型名称遵循model:tag的格式
    • 其中model可以具有可选的命名空间，例如example/model
      • 示例：orca-mini:3b-q4_1 和llama3:70b
    • tag是可选项，没有则默认为latest
      • 用来确定模型的特定版本

持续时间

  • 所有持续时间均以纳秒为单位返回

流式处理响应 / Streaming responses

  • 某些终端节点将响应作为 JSON 对象流式传输
  • 可以通过提供{"stream": false}这些终端节点来禁用流式处理。

模型生成回复

1
POST /api/generate
  • 使用提供的模型为给定Prompt生成响应
  • 这是一个流式处理终结点，因此将有一系列响应
  • 最终响应对象将包含来自请求的统计信息和其他数据

参数

  • model：（必需）模型名称
  • prompt：生成响应的提示
  • suffix：模型响应后的文本
  • images：（可选）base64 编码图像列表（对于多模态模型，例如llava)
（可选）高级参数：
  • format：返回响应的格式。目前唯一接受的值是json
  • options：Modelfile 文档中列出的其他模型参数，例如temperature
  • system：系统消息更改为（覆盖Modelfile)
  • template：要使用的提示模板（覆盖Modelfile)
  • context：从上一个请求返回的 context 参数 to /generate，这可用于保持较短的对话记忆
  • stream：如果是false，响应将作为单个响应对象返回，而不是对象流
  • raw：如果true，没有格式将应用于Prompt。如果您在对 API 的请求中指定了完整的模板化Prompt，则可以选择使用raw参数
  • keep_alive：控制模型在请求后加载到内存中的时间（默认值：5m)
JSON模式
  • 将format参数设置为json，可以启用JSON模式
  • 这个模式可以将模型的响应结构化为有效的JSON对象
  • 参阅下面的 JSON 模式示例

重要提示：

务必在prompt中指示模型使用JSON格式

否则模型产生的回答会包含大量空格

示例

流式请求 （Streaming）
Request
1
2
3
4
curl http://localhost:11434/api/generate -d '{
  "model": "llama3.2",
  "prompt": "Why is the sky blue?"
}'
Response
1
2
3
4
5
6
{
  "model": "llama3.2",
  "created_at": "2023-08-04T08:52:19.385406455-07:00",
  "response": "The",
  "done": false
}

流中的最终响应还包括有关生成的其他数据：

  • total_duration：生成响应所花费的时间

  • load_duration：加载模型所花费的时间（以纳秒为单位）

  • prompt_eval_count：提示中的token数

  • prompt_eval_duration：评估提示所花费的时间（以纳秒为单位）

  • eval_count：响应中的token数

  • eval_duration：生成响应所花费的时间（以纳秒为单位）

  • context：此响应中使用的对话的编码，可以在下一个请求中发送以保持对话记忆

  • response：如果响应是流式的，则为空，如果非流式响应，则此响应将包含完整的响应

  • 要计算以每秒令牌 （token/s） 为单位生成响应的速度，除以 eval_count / eval_duration * 10^9

1
2
3
4
5
6
7
8
9
10
11
12
13
{
  "model": "llama3.2",
  "created_at": "2023-08-04T19:22:45.499127Z",
  "response": "",
  "done": true,
  "context": [1, 2, 3],
  "total_duration": 10706818083,
  "load_duration": 6338219291,
  "prompt_eval_count": 26,
  "prompt_eval_duration": 130079000,
  "eval_count": 259,
  "eval_duration": 4232710000
}
非流式请求 （No Streaming）
Request
  • 当流式传输关闭时，可以在一个回复中收到响应
1
2
3
4
5
curl http://localhost:11434/api/generate -d '{
  "model": "llama3.2",
  "prompt": "Why is the sky blue?",
  "stream": false
}'
Response
  • 如果stream被设置为false，则Response将是单个 JSON 对象
1
2
3
4
5
6
7
8
9
10
11
12
13
{
  "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": 5043500667,
  "load_duration": 5025959,
  "prompt_eval_count": 26,
  "prompt_eval_duration": 325953000,
  "eval_count": 290,
  "eval_duration": 4709213000
}
带后缀的请求（with suffix）
Request
1
2
3
4
5
6
7
8
9
curl http://localhost:11434/api/generate -d '{
  "model": "codellama:code",
  "prompt": "def compute_gcd(a, b):",
  "suffix": "    return result",
  "options": {
    "temperature": 0
  },
  "stream": false
}'
Response
1
2
3
4
5
6
7
8
9
10
11
12
13
14
{
  "model": "codellama:code",
  "created_at": "2024-07-22T20:47:51.147561Z",
  "response": "\n  if a == 0:\n    return b\n  else:\n    return compute_gcd(b % a, a)\n\ndef compute_lcm(a, b):\n  result = (a * b) / compute_gcd(a, b)\n",
  "done": true,
  "done_reason": "stop",
  "context": [...],
  "total_duration": 1162761250,
  "load_duration": 6683708,
  "prompt_eval_count": 17,
  "prompt_eval_duration": 201222000,
  "eval_count": 63,
  "eval_duration": 953997000
}
生成请求（JSON 模式）

当format被设置为json时，输出将始终是格式正确的JSON对象

instruct 模型 用JSON格式回答

Request
1
2
3
4
5
6
curl http://localhost:11434/api/generate -d '{
  "model": "llama3.2",
  "prompt": "What color is the sky at different times of the day? Respond using JSON",
  "format": "json",
  "stream": false
}'
Response
1
2
3
4
5
6
7
8
9
10
11
12
13
{
  "model": "llama3.2",
  "created_at": "2023-11-09T21:07:55.186497Z",
  "response": "{\n\"morning\": {\n\"color\": \"blue\"\n},\n\"noon\": {\n\"color\": \"blue-gray\"\n},\n\"afternoon\": {\n\"color\": \"warm gray\"\n},\n\"evening\": {\n\"color\": \"orange\"\n}\n}\n",
  "done": true,
  "context": [1, 2, 3],
  "total_duration": 4648158584,
  "load_duration": 4071084,
  "prompt_eval_count": 36,
  "prompt_eval_duration": 439038000,
  "eval_count": 180,
  "eval_duration": 4196918000
}
  • Response的值将是包含JSON的字符串，类似于：
1
2
3
4
5
6
7
8
9
10
11
12
13
14
{
  "morning": {
    "color": "blue"
  },
  "noon": {
    "color": "blue-gray"
  },
  "afternoon": {
    "color": "warm gray"
  },
  "evening": {
    "color": "orange"
  }
}
请求（带图片）
  • 要将图像提交到例如llava或bakllava的多模型模型，需要提供base64编码的images
Request
1
2
3
4
5
6
curl http://localhost:11434/api/generate -d '{
  "model": "llava",
  "prompt":"What is in this picture?",
  "stream": false,
  "images": ["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"]
}'
Response
1
2
3
4
5
6
7
8
9
10
11
12
13
{
  "model": "llava",
  "created_at": "2023-11-03T15:36:02.583064Z",
  "response": "A happy cartoon character, which is cute and cheerful.",
  "done": true,
  "context": [1, 2, 3],
  "total_duration": 2938432250,
  "load_duration": 2559292,
  "prompt_eval_count": 1,
  "prompt_eval_duration": 2195557000,
  "eval_count": 44,
  "eval_duration": 736432000
}
请求（Raw 模式）
  • 在某些情况下，您可能希望绕过模板系统（templating system）并提供完整提示（provide a full prompt）
  • 在这种情况下，您可以使用raw 参数来禁用模板（disable templating）
  • 另请注意，raw 模式不会返回上下文（not return a context）
Request
1
2
3
4
5
6
curl http://localhost:11434/api/generate -d '{
  "model": "mistral",
  "prompt": "[INST] why is the sky blue? [/INST]",
  "raw": true,
  "stream": false
}'
请求（可重现的输出）
  • 对于可重现的输出，将seed设置为一个数字
Request
1
2
3
4
5
6
7
curl http://localhost:11434/api/generate -d '{
  "model": "mistral",
  "prompt": "Why is the sky blue?",
  "options": {
    "seed": 123
  }
}'
Response
1
2
3
4
5
6
7
8
9
10
11
12
{
  "model": "mistral",
  "created_at": "2023-11-03T15:36:02.583064Z",
  "response": " The sky appears blue because of a phenomenon called Rayleigh scattering.",
  "done": true,
  "total_duration": 8493852375,
  "load_duration": 6589624375,
  "prompt_eval_count": 14,
  "prompt_eval_duration": 119039000,
  "eval_count": 110,
  "eval_duration": 1779061000
}
生成请求（带选项）
  • 如果想要在模型运行时设置自定义选项，而不是使用Modelfile，可以使用options参数来实现
  • 以下示例设置了所有可用的选项
  • 实际使用时，可以单独设置其中任何一个选项，省略不想覆盖的选项
Request
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
curl http://localhost:11434/api/generate -d '{
  "model": "llama3.2",
  "prompt": "Why is the sky blue?",
  "stream": false,
  "options": {
    "num_keep": 5,
    "seed": 42,
    "num_predict": 100,
    "top_k": 20,
    "top_p": 0.9,
    "min_p": 0.0,
    "tfs_z": 0.5,
    "typical_p": 0.7,
    "repeat_last_n": 33,
    "temperature": 0.8,
    "repeat_penalty": 1.2,
    "presence_penalty": 1.5,
    "frequency_penalty": 1.0,
    "mirostat": 1,
    "mirostat_tau": 0.8,
    "mirostat_eta": 0.6,
    "penalize_newline": true,
    "stop": ["\n", "user:"],
    "numa": false,
    "num_ctx": 1024,
    "num_batch": 2,
    "num_gpu": 1,
    "main_gpu": 0,
    "low_vram": false,
    "f16_kv": true,
    "vocab_only": false,
    "use_mmap": true,
    "use_mlock": false,
    "num_thread": 8
  }
}'
Response
1
2
3
4
5
6
7
8
9
10
11
12
13
{
  "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
}
加载模型
  • 如果提供了空的prompt，则模型将被加载到内存中
Request
1
2
3
curl http://localhost:11434/api/generate -d '{
  "model": "llama3.2"
}'
Response

返回单个JSON对象

1
2
3
4
5
6
{
  "model": "llama3.2",
  "created_at": "2023-12-18T19:52:07.071755Z",
  "response": "",
  "done": true
}
卸载模型
  • 如果提供的空的prompt，且keep_alive参数被设置为0，则将从内存中卸载模型
Request
1
2
3
4
curl http://localhost:11434/api/generate -d '{
  "model": "llama3.2",
  "keep_alive": 0
}'
Response

返回单个JSON对象

1
2
3
4
5
6
7
{
  "model": "llama3.2",
  "created_at": "2024-09-12T03:54:03.516566Z",
  "response": "",
  "done": true,
  "done_reason": "unload"
}

与模型完成对话

1
POST /api/chat
  • 用提供的模型在对话中生成下一条消息
  • 这是一个流式处理终结点，所以会生成一系列响应
  • 可以通过使用"stream: false"来禁用流式传输
  • 最终响应对象将包含来自request的统计信息和其他数据

参数

  • model：（必需）模型名称
  • messages：聊天的消息，可用于保存聊天记录
  • tools：模型要使用的工具（如果支持）。需要把stream设置为false

message对象具有以下字段：

  • role：消息的角色，可以是system 、user 、assistant 或tool
  • content：消息的内容
  • images（可选）：要包含在消息中的图像列表（对于多模态模型，例如llava)
  • tool_calls（可选）：模型要使用的工具列表

（可选）高级参数：

  • format：返回响应的格式。目前唯一接受的值是json
  • options：Modelfile 文档中列出的其他模型参数，例如temperature
  • stream：如果设置为false，响应将作为单个响应对象返回，而不是对象流
  • keep_alive：控制模型在请求后加载到内存中的时间（默认值：5m)

示例

对话请求（流式处理）
Request
  • 发送包含流式响应的聊天消息
1
2
3
4
5
6
7
8
9
curl http://localhost:11434/api/chat -d '{
  "model": "llama3.2",
  "messages": [
    {
      "role": "user",
      "content": "why is the sky blue?"
    }
  ]
}'
Response
  • 返回JSON对象流
1
2
3
4
5
6
7
8
9
10
{
  "model": "llama3.2",
  "created_at": "2023-08-04T08:52:19.385406455-07:00",
  "message": {
    "role": "assistant",
    "content": "The",
    "images": null
  },
  "done": false
}
  • 最终Response
1
2
3
4
5
6
7
8
9
10
11
{
  "model": "llama3.2",
  "created_at": "2023-08-04T19:22:45.499127Z",
  "done": true,
  "total_duration": 4883583458,
  "load_duration": 1334875,
  "prompt_eval_count": 26,
  "prompt_eval_duration": 342546000,
  "eval_count": 282,
  "eval_duration": 4535599000
}
对话请求（无流式处理）
Request
1
2
3
4
5
6
7
8
9
10
curl http://localhost:11434/api/chat -d '{
  "model": "llama3.2",
  "messages": [
    {
      "role": "user",
      "content": "why is the sky blue?"
    }
  ],
  "stream": false
}'

新增参数stream和设置false

Response
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
{
  "model": "llama3.2",
  "created_at": "2023-12-12T14:13:43.416799Z",
  "message": {
    "role": "assistant",
    "content": "Hello! How are you today?"
  },
  "done": true,
  "total_duration": 5191566416,
  "load_duration": 2154458,
  "prompt_eval_count": 26,
  "prompt_eval_duration": 383809000,
  "eval_count": 298,
  "eval_duration": 4799921000
}
对话请求（带历史记录）
  • 发送包含对话历史记录的聊天消息
  • 这个方法同样可以用来进行**多发（multi-shot）或思维链提示（chain-of-thought prompting）**的对话
Request
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
curl http://localhost:11434/api/chat -d '{
  "model": "llama3.2",
  "messages": [
    {
      "role": "user",
      "content": "why is the sky blue?"
    },
    {
      "role": "assistant",
      "content": "due to rayleigh scattering."
    },
    {
      "role": "user",
      "content": "how is that different than mie scattering?"
    }
  ]
}'
Response
  • 返回JSON对象流
1
2
3
4
5
6
7
8
9
{
  "model": "llama3.2",
  "created_at": "2023-08-04T08:52:19.385406455-07:00",
  "message": {
    "role": "assistant",
    "content": "The"
  },
  "done": false
}
  • 最终回应
1
2
3
4
5
6
7
8
9
10
11
{
  "model": "llama3.2",
  "created_at": "2023-08-04T19:22:45.499127Z",
  "done": true,
  "total_duration": 8113331500,
  "load_duration": 6396458,
  "prompt_eval_count": 61,
  "prompt_eval_duration": 398801000,
  "eval_count": 468,
  "eval_duration": 7701267000
}
对话请求（带图片）
Request
  • 发送带有图像的聊天消息
  • 图像应该以数组（array）形式提供
  • 单个图像以Base64编码
1
2
3
4
5
6
7
8
9
10
curl http://localhost:11434/api/chat -d '{
  "model": "llava",
  "messages": [
    {
      "role": "user",
      "content": "what is in this image?",
      "images": ["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"]
    }
  ]
}'
Response
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
{
  "model": "llava",
  "created_at": "2023-12-13T22:42:50.203334Z",
  "message": {
    "role": "assistant",
    "content": " The image features a cute, little pig with an angry facial expression. It's wearing a heart on its shirt and is waving in the air. This scene appears to be part of a drawing or sketching project.",
    "images": null
  },
  "done": true,
  "total_duration": 1668506709,
  "load_duration": 1986209,
  "prompt_eval_count": 26,
  "prompt_eval_duration": 359682000,
  "eval_count": 83,
  "eval_duration": 1303285000
}
对话请求（可重现的输出）
Request
1
2
3
4
5
6
7
8
9
10
11
12
13
curl http://localhost:11434/api/chat -d '{
  "model": "llama3.2",
  "messages": [
    {
      "role": "user",
      "content": "Hello!"
    }
  ],
  "options": {
    "seed": 101,
    "temperature": 0
  }
}'
Request
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
{
  "model": "llama3.2",
  "created_at": "2023-12-12T14:13:43.416799Z",
  "message": {
    "role": "assistant",
    "content": "Hello! How are you today?"
  },
  "done": true,
  "total_duration": 5191566416,
  "load_duration": 2154458,
  "prompt_eval_count": 26,
  "prompt_eval_duration": 383809000,
  "eval_count": 298,
  "eval_duration": 4799921000
}
对话请求（带工具）
Request
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
curl http://localhost:11434/api/chat -d '{
  "model": "llama3.2",
  "messages": [
    {
      "role": "user",
      "content": "What is the weather today in Paris?"
    }
  ],
  "stream": false,
  "tools": [
    {
      "type": "function",
      "function": {
        "name": "get_current_weather",
        "description": "Get the current weather for a location",
        "parameters": {
          "type": "object",
          "properties": {
            "location": {
              "type": "string",
              "description": "The location to get the weather for, e.g. San Francisco, CA"
            },
            "format": {
              "type": "string",
              "description": "The format to return the weather in, e.g. 'celsius' or 'fahrenheit'",
              "enum": ["celsius", "fahrenheit"]
            }
          },
          "required": ["location", "format"]
        }
      }
    }
  ]
}'
Response
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
{
  "model": "llama3.2",
  "created_at": "2024-07-22T20:33:28.123648Z",
  "message": {
    "role": "assistant",
    "content": "",
    "tool_calls": [
      {
        "function": {
          "name": "get_current_weather",
          "arguments": {
            "format": "celsius",
            "location": "Paris, FR"
          }
        }
      }
    ]
  },
  "done_reason": "stop",
  "done": true,
  "total_duration": 885095291,
  "load_duration": 3753500,
  "prompt_eval_count": 122,
  "prompt_eval_duration": 328493000,
  "eval_count": 33,
  "eval_duration": 552222000
}

加载模型

  • 如果提供的messages数组为空，则模型将被加载到内存中
Request
1
2
3
4
curl http://localhost:11434/api/chat -d '{
  "model": "llama3.2",
  "messages": []
}'
Response
1
2
3
4
5
6
7
8
9
10
{
  "model": "llama3.2",
  "created_at":"2024-09-12T21:17:29.110811Z",
  "message": {
    "role": "assistant",
    "content": ""
  },
  "done_reason": "load",
  "done": true
}

卸载模型

  • 如果提供的空的messages数组，且keep_alive参数被设置为0，则将从内存中卸载模型
Request
1
2
3
4
5
curl http://localhost:11434/api/chat -d '{
  "model": "llama3.2",
  "messages": [],
  "keep_alive": 0
}'
Response

返回单个JSON对象

1
2
3
4
5
6
7
8
9
10
{
  "model": "llama3.2",
  "created_at":"2024-09-12T21:33:17.547535Z",
  "message": {
    "role": "assistant",
    "content": ""
  },
  "done_reason": "unload",
  "done": true
}

创建模型

1
POST /api/create
  • 从 Modelfile 创建模型
  • 远程创建的要求：建议设置modelfile为 Modelfile 的内容，而不仅仅是设置path
  • 远程模型创建还必须使用创建 Blob 和response中指示的path的值与服务器显式创建任何文件 blob（any file blobs）、fields字段（如 FROM和 ADAPTER）

参数（Parameters）

  • name：要创建的模型的名称
  • modelfile（可选）：Modelfile 的内容
  • stream：（可选）如果设置为false，响应将作为单个响应对象（a single response object）而不是对象流（a stream of objects）返回
  • path（可选）：Modelfile 的路径

示例

创建新模型

  • 从Modelfile创建新模型
Request
1
2
3
4
curl http://localhost:11434/api/create -d '{
  "name": "mario",
  "modelfile": "FROM llama3\nSYSTEM You are mario from Super Mario Bros."
}'
Response
  • 响应会返回JSON对象流
  • 最终的JSON对象显示为一个"status": "success"
1
2
3
4
5
6
7
8
9
10
11
{"status":"reading model metadata"}
{"status":"creating system layer"}
{"status":"using already created layer sha256:22f7f8ef5f4c791c1b03d7eb414399294764d7cc82c7e94aa81a1feb80a983a2"}
{"status":"using already created layer sha256:8c17c2ebb0ea011be9981cc3922db8ca8fa61e828c5d3f44cb6ae342bf80460b"}
{"status":"using already created layer sha256:7c23fb36d80141c4ab8cdbb61ee4790102ebd2bf7aeff414453177d4f2110e5d"}
{"status":"using already created layer sha256:2e0493f67d0c8c9c68a8aeacdf6a38a2151cb3c4c1d42accf296e19810527988"}
{"status":"using already created layer sha256:2759286baa875dc22de5394b4a925701b1896a7e3f8e53275c36f75a877a82c9"}
{"status":"writing layer sha256:df30045fe90f0d750db82a058109cecd6d4de9c90a3d75b19c09e5f64580bb42"}
{"status":"writing layer sha256:f18a68eb09bf925bb1b669490407c1b1251c5db98dc4d3d81f3088498ea55690"}
{"status":"writing manifest"}
{"status":"success"}

自分の插入内容：什么是Blob

  • Blob（Binary Large Object）是一种用于表示二进制数据的类型
  • Blob对象可以用来表示大量数据，如图像、视频、音频文件等
  • Blob对象是只读的，但可以通过Blob构造函数从其他数据源（如数组缓冲区、字符串等）创建新的Blob对象

主要特点

  • 二进制数据：Blob对象主要用于处理二进制数据，可以存储和操作大量数据。
  • 只读：Blob对象是不可变的，一旦创建就不能修改。
  • 类型：Blob对象有一个type属性，表示数据的MIME类型，默认值为""（空字符串）。
  • 大小：Blob对象有一个size属性，表示数据的字节大小。

常见用途

  • 文件上传：将文件转换为Blob对象，然后通过FormData对象发送到服务器。
  • 文件下载：从服务器接收二进制数据，创建Blob对象，然后生成一个下载链接供用户下载。
  • 多媒体处理：处理图像、视频、音频等多媒体数据。
  • 本地存储：将Blob对象存储在浏览器的IndexedDB中，用于离线应用

创建和使用Blob对象

创建Blob对象
1
2
3
4
5
6
7
8
9
10
11
12
13
14
// 从字符串创建Blob对象
const text = "Hello, World!";
const blob = new Blob([text], { type: "text/plain" });

// 从数组缓冲区创建Blob对象
const arrayBuffer = new ArrayBuffer(8);
const blob = new Blob([arrayBuffer], { type: "application/octet-stream" });

// 从文件输入创建Blob对象
const fileInput = document.querySelector('input[type="file"]');
fileInput.addEventListener('change', (event) => {
  const file = event.target.files[0];
  const blob = new Blob([file], { type: file.type });
});
读取Blob对象
1
2
3
4
5
6
7
8
9
// 使用FileReader读取Blob对象
const reader = new FileReader();
reader.onload = function(event) {
  const content = event.target.result;
  console.log(content); // 输出Blob对象的内容
};
reader.readAsText(blob); // 读取为文本
// reader.readAsDataURL(blob); // 读取为Data URL
// reader.readAsArrayBuffer(blob); // 读取为数组缓冲区
发送Blob对象
1
2
3
4
5
6
7
8
9
10
11
// 使用FormData发送Blob对象
const formData = new FormData();
formData.append('file', blob, 'filename.txt');

fetch('/upload', {
  method: 'POST',
  body: formData
})
.then(response => response.json())
.then(data => console.log(data))
.catch(error => console.error('Error:', error));
下载Blob对象
1
2
3
4
5
6
7
// 生成下载链接
const link = document.createElement('a');
link.href = URL.createObjectURL(blob);
link.download = 'filename.txt';
document.body.appendChild(link);
link.click();
document.body.removeChild(link);
文件上传
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
<!DOCTYPE html>
<html lang="en">
<head>
  <meta charset="UTF-8">
  <meta name="viewport" content="width=device-width, initial-scale=1.0">
  <title>File Upload</title>
</head>
<body>
  <input type="file" id="fileInput">
  <button id="uploadButton">Upload</button>

  <script>
    const fileInput = document.getElementById('fileInput');
    const uploadButton = document.getElementById('uploadButton');

    uploadButton.addEventListener('click', () => {
      const file = fileInput.files[0];
      if (file) {
        const formData = new FormData();
        formData.append('file', file, file.name);

        fetch('/upload', {
          method: 'POST',
          body: formData
        })
        .then(response => response.json())
        .then(data => console.log(data))
        .catch(error => console.error('Error:', error));
      }
    });
  </script>
</body>
</html>
文件下载
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
<!DOCTYPE html>
<html lang="en">
<head>
  <meta charset="UTF-8">
  <meta name="viewport" content="width=device-width, initial-scale=1.0">
  <title>File Download</title>
</head>
<body>
  <button id="downloadButton">Download</button>

  <script>
    const downloadButton = document.getElementById('downloadButton');

    downloadButton.addEventListener('click', () => {
      fetch('/download')
        .then(response => response.blob())
        .then(blob => {
          const link = document.createElement('a');
          link.href = URL.createObjectURL(blob);
          link.download = 'filename.txt';
          document.body.appendChild(link);
          link.click();
          document.body.removeChild(link);
        })
        .catch(error => console.error('Error:', error));
    });
  </script>
</body>
</html>

检查Blob是否存在

1
HEAD /api/blobs/:digest
  • 确保服务器上存在用于FROM或ADAPTER字段的文件blob
  • 检查的是自己的Ollama服务器，而不是Ollama.ai

查询参数

  • digest：blob 的 SHA256 摘要

示例

Request
1
curl -I http://localhost:11434/api/blobs/sha256:29fdb92e57cf0827ded04ae6461b5931d01fa595843f55d36f5b275a52087dd2
Response
  • 如果blob存在，则返回状态码为200 OK
  • blob不存在，则返回404 Not Found

创建Blob

1
POST /api/blobs/:digest
  • 用服务器的文件创建Blob
  • 然后返回服务器文件路径

查询参数

  • digest：文件的预期 SHA256 摘要

示例

Request
1
curl -T model.bin -X POST http://localhost:11434/api/blobs/sha256:29fdb92e57cf0827ded04ae6461b5931d01fa595843f55d36f5b275a52087dd2
Response
  • 如果 blob 创建成功，则返回201 Created
  • 如果使用的摘要不是预期的，则返回 400 Bad Request

列出本地模型

1
GET /api/tags

列出本地可用的模型

示例

Request

1
curl http://localhost:11434/api/tags

Response

  • 将返回单个JSON对象
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
{
  "models": [
    {
      "name": "codellama:13b",
      "modified_at": "2023-11-04T14:56:49.277302595-07:00",
      "size": 7365960935,
      "digest": "9f438cb9cd581fc025612d27f7c1a6669ff83a8bb0ed86c94fcf4c5440555697",
      "details": {
        "format": "gguf",
        "family": "llama",
        "families": null,
        "parameter_size": "13B",
        "quantization_level": "Q4_0"
      }
    },
    {
      "name": "llama3:latest",
      "modified_at": "2023-12-07T09:32:18.757212583-08:00",
      "size": 3825819519,
      "digest": "fe938a131f40e6f6d40083c9f0f430a515233eb2edaa6d72eb85c50d64f2300e",
      "details": {
        "format": "gguf",
        "family": "llama",
        "families": null,
        "parameter_size": "7B",
        "quantization_level": "Q4_0"
      }
    }
  ]
}

显示模型信息

1
POST /api/show
  • 显示有关模型的信息
  • 包括详细信息、模板文件、模板、参数、许可证、系统提示符

参数

  • name：要显示的模型的名称
  • verbose：（可选）如果设置为true，返回详细响应字段的完整数据

示例

Request
1
2
3
curl http://localhost:11434/api/show -d '{
  "name": "llama3.2"
}'
Response
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
{
  "modelfile": "# Modelfile generated by \"ollama show\"\n# To build a new Modelfile based on this one, replace the FROM line with:\n# FROM llava:latest\n\nFROM /Users/matt/.ollama/models/blobs/sha256:200765e1283640ffbd013184bf496e261032fa75b99498a9613be4e94d63ad52\nTEMPLATE \"\"\"{{ .System }}\nUSER: {{ .Prompt }}\nASSISTANT: \"\"\"\nPARAMETER num_ctx 4096\nPARAMETER stop \"\u003c/s\u003e\"\nPARAMETER stop \"USER:\"\nPARAMETER stop \"ASSISTANT:\"",
  "parameters": "num_keep                       24\nstop                           \"<|start_header_id|>\"\nstop                           \"<|end_header_id|>\"\nstop                           \"<|eot_id|>\"",
  "template": "{{ if .System }}<|start_header_id|>system<|end_header_id|>\n\n{{ .System }}<|eot_id|>{{ end }}{{ if .Prompt }}<|start_header_id|>user<|end_header_id|>\n\n{{ .Prompt }}<|eot_id|>{{ end }}<|start_header_id|>assistant<|end_header_id|>\n\n{{ .Response }}<|eot_id|>",
  "details": {
    "parent_model": "",
    "format": "gguf",
    "family": "llama",
    "families": [
      "llama"
    ],
    "parameter_size": "8.0B",
    "quantization_level": "Q4_0"
  },
  "model_info": {
    "general.architecture": "llama",
    "general.file_type": 2,
    "general.parameter_count": 8030261248,
    "general.quantization_version": 2,
    "llama.attention.head_count": 32,
    "llama.attention.head_count_kv": 8,
    "llama.attention.layer_norm_rms_epsilon": 0.00001,
    "llama.block_count": 32,
    "llama.context_length": 8192,
    "llama.embedding_length": 4096,
    "llama.feed_forward_length": 14336,
    "llama.rope.dimension_count": 128,
    "llama.rope.freq_base": 500000,
    "llama.vocab_size": 128256,
    "tokenizer.ggml.bos_token_id": 128000,
    "tokenizer.ggml.eos_token_id": 128009,
    "tokenizer.ggml.merges": [],            // populates if `verbose=true`
    "tokenizer.ggml.model": "gpt2",
    "tokenizer.ggml.pre": "llama-bpe",
    "tokenizer.ggml.token_type": [],        // populates if `verbose=true`
    "tokenizer.ggml.tokens": []             // populates if `verbose=true`
  }
}

复制模型

1
POST /api/copy
  • 复制模型，从现有模型创建具有其他名称的模型

例子

Request

1
2
3
4
curl http://localhost:11434/api/copy -d '{
  "source": "llama3.2",
  "destination": "llama3-backup"
}'

Response

  • 如果成功，则返回 200 OK
  • 如果源模型不存在，则返回 404 Not Found

删除模型

1
DELETE /api/delete
  • 删除模型及其数据

参数

  • name：要删除的型号名称

示例

Request

1
2
3
curl -X DELETE http://localhost:11434/api/delete -d '{
  "name": "llama3:13b"
}'

Response

  • 如果成功，则返回 200 OK
  • 如果要删除的模型不存在，则返回 404 Not Found

拉取模型

1
POST /api/pull
  • 从 ollama 库下载模型
  • 取消的拉取将从上次中断的位置继续
  • 多个调用将共享相同的下载进度

参数

  • name：要拉取的模型的名称
  • insecure：（可选）允许与库建立不安全的连接。仅在开发过程中从自己的库中提取时才使用此项。
  • stream：（可选）如果响应将作为单个响应对象而不是对象流返回false

示例

Request

1
2
3
curl http://localhost:11434/api/pull -d '{
  "name": "llama3.2"
}'

Response

  • 如果stream未指定 / 被设置为true，则返回JSON对象流
  • 第一个对象是清单：
1
2
3
{
  "status": "pulling manifest"
}
  • 然后是一系列的下载响应
  • 在任何下载完成之前，completed密钥可能不包含在内
  • 要下载的文件数取决于清单中指定的层数
1
2
3
4
5
6
{
  "status": "downloading digestname",
  "digest": "digestname",
  "total": 2142590208,
  "completed": 241970
}
  • 下载所有文件后，最终响应为：
1
2
3
4
5
6
7
8
9
10
11
12
{
    "status": "verifying sha256 digest"
}
{
    "status": "writing manifest"
}
{
    "status": "removing any unused layers"
}
{
    "status": "success"
}
  • 如果将stream设置为false，则响应为单个JSON对象
1
2
3
{
  "status": "success"
}

推送模型

1
POST /api/push
  • 将模型上传到模型库。需要先注册 ollama.ai并添加公钥

参数

  • name：要推送的模型名称，形式为<namespace>/<model>:<tag>
  • insecure：（可选）允许与库建立不安全的连接。仅在开发期间推送到库时使用此选项。
  • stream：（可选）如果设置为false，响应将作为单个响应对象而不是对象流返回

示例

Request

1
2
3
curl http://localhost:11434/api/push -d '{
  "name": "mattw/pygmalion:latest"
}'

Response

  • 如果stream未指定 / 被设置为true，则返回JSON对象流
1
{ "status": "retrieving manifest" }
  • 然后是一系列上传响应：
1
2
3
4
5
{
  "status": "starting upload",
  "digest": "sha256:bc07c81de745696fdf5afca05e065818a8149fb0c77266fb584d9b2cba3711ab",
  "total": 1928429856
}
  • 最后，上传完成
1
2
{"status":"pushing manifest"}
{"status":"success"}

生成嵌入（Generate Embeddings）

1
POST /api/embed
  • 从模型生成嵌入

参数

  • model：要从中生成嵌入的模型的名称
  • input：要为其生成嵌入的文本或文本列表

高级参数

  • truncate：截断每个输入的结尾以适合上下文长度。如果参数值为false且上下文长度超出，则返回错误。默认为true
  • options：Modelfile 文档中列出的其他模型参数，例如temperature
  • keep_alive：控制模型在请求后加载到内存中的时间（默认值：5m)

示例

Request

1
2
3
4
curl http://localhost:11434/api/embed -d '{
  "model": "all-minilm",
  "input": "Why is the sky blue?"
}'

Response

1
2
3
4
5
6
7
8
9
10
{
  "model": "all-minilm",
  "embeddings": [[
    0.010071029, -0.0017594862, 0.05007221, 0.04692972, 0.054916814,
    0.008599704, 0.105441414, -0.025878139, 0.12958129, 0.031952348
  ]],
  "total_duration": 14143917,
  "load_duration": 1019500,
  "prompt_eval_count": 8
}

Request（多路输入）

1
2
3
4
curl http://localhost:11434/api/embed -d '{
  "model": "all-minilm",
  "input": ["Why is the sky blue?", "Why is the grass green?"]
}'

Response（多路输入的）

1
2
3
4
5
6
7
8
9
10
{
  "model": "all-minilm",
  "embeddings": [[
    0.010071029, -0.0017594862, 0.05007221, 0.04692972, 0.054916814,
    0.008599704, 0.105441414, -0.025878139, 0.12958129, 0.031952348
  ],[
    -0.0098027075, 0.06042469, 0.025257962, -0.006364387, 0.07272725,
    0.017194884, 0.09032035, -0.051705178, 0.09951512, 0.09072481
  ]]
}

列出正在运行的模型

1
GET /api/ps
  • 列出当前加载到内存中的模型

示例

Request

1
curl http://localhost:11434/api/ps

Response

  • 返回单个JSON对象
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
{
  "models": [
    {
      "name": "mistral:latest",
      "model": "mistral:latest",
      "size": 5137025024,
      "digest": "2ae6f6dd7a3dd734790bbbf58b8909a606e0e7e97e94b7604e0aa7ae4490e6d8",
      "details": {
        "parent_model": "",
        "format": "gguf",
        "family": "llama",
        "families": [
          "llama"
        ],
        "parameter_size": "7.2B",
        "quantization_level": "Q4_0"
      },
      "expires_at": "2024-06-04T14:38:31.83753-07:00",
      "size_vram": 5137025024
    }
  ]
}
  • Copyrights © 2024-2025 brocademaple
  • 访问人数: | 浏览次数:

      请我喝杯咖啡吧~

      支付宝
      微信