api.md 28.9 KB
Newer Older
1
2
# API

3
4
5
## Endpoints

- [Generate a completion](#generate-a-completion)
6
- [Generate a chat completion](#generate-a-chat-completion)
Matt Williams's avatar
Matt Williams committed
7
8
9
10
11
12
13
14
15
- [Create a Model](#create-a-model)
- [List Local Models](#list-local-models)
- [Show Model Information](#show-model-information)
- [Copy a Model](#copy-a-model)
- [Delete a Model](#delete-a-model)
- [Pull a Model](#pull-a-model)
- [Push a Model](#push-a-model)
- [Generate Embeddings](#generate-embeddings)

16
## Conventions
Matt Williams's avatar
Matt Williams committed
17

18
### Model names
Matt Williams's avatar
Matt Williams committed
19

Matt Williams's avatar
Matt Williams committed
20
Model names follow a `model:tag` format. Some examples are `orca-mini:3b-q4_1` and `llama2:70b`. The tag is optional and, if not provided, will default to `latest`. The tag is used to identify a specific version.
Matt Williams's avatar
Matt Williams committed
21
22
23

### Durations

24
All durations are returned in nanoseconds.
Matt Williams's avatar
Matt Williams committed
25

26
27
### Streaming responses

Bruce MacDonald's avatar
Bruce MacDonald committed
28
Certain endpoints stream responses as JSON objects.
29

30
## Generate a completion
Matt Williams's avatar
Matt Williams committed
31

Matt Williams's avatar
Matt Williams committed
32
```shell
33
34
POST /api/generate
```
35

Bruce MacDonald's avatar
Bruce MacDonald committed
36
Generate a response for a given prompt with a provided model. This is a streaming endpoint, so there will be a series of responses. The final response object will include statistics and additional data from the request.
Matt Williams's avatar
Matt Williams committed
37

38
### Parameters
Matt Williams's avatar
Matt Williams committed
39

40
41
- `model`: (required) the [model name](#model-names)
- `prompt`: the prompt to generate a response for
42
- `images`: a list of base64-encoded images (for multimodal models such as `llava`)
Matt Williams's avatar
Matt Williams committed
43

44
Advanced parameters (optional):
Matt Williams's avatar
Matt Williams committed
45

46
- `format`: the format to return a response in. Currently the only accepted value is `json`
47
- `options`: additional model parameters listed in the documentation for the [Modelfile](./modelfile.md#valid-parameters-and-values) such as `temperature`
48
- `system`: system message to (overrides what is defined in the `Modelfile`)
49
50
- `template`: the full prompt or prompt template (overrides what is defined in the `Modelfile`)
- `context`: the context parameter returned from a previous request to `/generate`, this can be used to keep a short conversational memory
51
- `stream`: if `false` the response will be returned as a single response object, rather than a stream of objects
Bruce MacDonald's avatar
Bruce MacDonald committed
52
- `raw`: if `true` no formatting will be applied to the prompt. You may choose to use the `raw` parameter if you are specifying a full templated prompt in your request to the API.
53

54
55
### JSON mode

56
57
58
Enable JSON mode by setting the `format` parameter to `json`. This will structure the response as valid JSON. See the JSON mode [example](#request-json-mode) below.

> Note: it's important to instruct the model to use JSON in the `prompt`. Otherwise, the model may generate large amounts whitespace.
59

60
61
### Examples

Jeffrey Morgan's avatar
Jeffrey Morgan committed
62
#### Request
Matt Williams's avatar
Matt Williams committed
63

Matt Williams's avatar
Matt Williams committed
64
```shell
65
curl http://localhost:11434/api/generate -d '{
66
  "model": "llama2",
67
68
69
  "prompt": "Why is the sky blue?"
}'
```
Matt Williams's avatar
Matt Williams committed
70

71
#### Response
Matt Williams's avatar
Matt Williams committed
72

73
A stream of JSON objects is returned:
Matt Williams's avatar
Matt Williams committed
74

75
```json
Matt Williams's avatar
Matt Williams committed
76
{
77
  "model": "llama2",
78
79
  "created_at": "2023-08-04T08:52:19.385406455-07:00",
  "response": "The",
Matt Williams's avatar
Matt Williams committed
80
81
82
83
  "done": false
}
```

84
The final response in the stream also includes additional data about the generation:
Matt Williams's avatar
Matt Williams committed
85

86
87
88
89
90
91
92
93
- `total_duration`: time spent generating the response
- `load_duration`: time spent in nanoseconds loading the model
- `sample_count`: number of samples generated
- `sample_duration`: time spent generating samples
- `prompt_eval_count`: number of tokens in the prompt
- `prompt_eval_duration`: time spent in nanoseconds evaluating the prompt
- `eval_count`: number of tokens the response
- `eval_duration`: time in nanoseconds spent generating the response
94
- `context`: an encoding of the conversation used in this response, this can be sent in the next request to keep a conversational memory
95
- `response`: empty if the response was streamed, if not streamed, this will contain the full response
96
97
98

To calculate how fast the response is generated in tokens per second (token/s), divide `eval_count` / `eval_duration`.

99
```json
Matt Williams's avatar
Matt Williams committed
100
{
101
  "model": "llama2",
102
  "created_at": "2023-08-04T19:22:45.499127Z",
103
  "response": "",
Bruce MacDonald's avatar
Bruce MacDonald committed
104
  "context": [1, 2, 3],
105
106
107
108
109
110
111
112
113
114
  "done": true,
  "total_duration": 5589157167,
  "load_duration": 3013701500,
  "sample_count": 114,
  "sample_duration": 81442000,
  "prompt_eval_count": 46,
  "prompt_eval_duration": 1160282000,
  "eval_count": 113,
  "eval_duration": 1325948000
}
Matt Williams's avatar
Matt Williams committed
115
116
```

117
#### Request (No streaming)
118

Bruce MacDonald's avatar
Bruce MacDonald committed
119
120
A response can be recieved in one reply when streaming is off.

121
```shell
122
curl http://localhost:11434/api/generate -d '{
123
  "model": "llama2",
124
125
126
127
128
129
130
  "prompt": "Why is the sky blue?",
  "stream": false
}'
```

#### Response

131
132
133
134
If `stream` is set to `false`, the response will be a single JSON object:

```json
{
135
  "model": "llama2",
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
  "created_at": "2023-08-04T19:22:45.499127Z",
  "response": "The sky is blue because it is the color of the sky.",
  "context": [1, 2, 3],
  "done": true,
  "total_duration": 5589157167,
  "load_duration": 3013701500,
  "sample_count": 114,
  "sample_duration": 81442000,
  "prompt_eval_count": 46,
  "prompt_eval_duration": 1160282000,
  "eval_count": 13,
  "eval_duration": 1325948000
}
```

151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
#### Request (with images)

To submit images to multimodal models such as `llava` or `bakllava`, provide a list of base64-encoded `images`:

```shell
curl http://localhost:11434/api/generate -d '{
  "model": "llava",
  "prompt":"What is in this picture?",
  "stream": false,
  "images": ["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"]
}'
```

#### Response

```
{
  "model": "llava",
  "created_at": "2023-11-03T15:36:02.583064Z",
  "response": "A happy cartoon character, which is cute and cheerful.",
  "context": [1, 2, 3],
  "done": true,
  "total_duration": 14648695333,
  "load_duration": 3302671417,
  "prompt_eval_count": 14,
  "prompt_eval_duration": 286243000,
  "eval_count": 129,
  "eval_duration": 10931424000
}
```

Bruce MacDonald's avatar
Bruce MacDonald committed
182
#### Request (Raw Mode)
183

Bruce MacDonald's avatar
Bruce MacDonald committed
184
In some cases you may wish to bypass the templating system and provide a full prompt. In this case, you can use the `raw` parameter to disable formatting.
185
186

```shell
187
curl http://localhost:11434/api/generate -d '{
188
189
190
191
192
193
194
195
196
197
198
199
200
201
  "model": "mistral",
  "prompt": "[INST] why is the sky blue? [/INST]",
  "raw": true,
  "stream": false
}'
```

#### Response

```json
{
  "model": "mistral",
  "created_at": "2023-11-03T15:36:02.583064Z",
  "response": " The sky appears blue because of a phenomenon called Rayleigh scattering.",
Bruce MacDonald's avatar
Bruce MacDonald committed
202
  "context": [1, 2, 3],
203
204
205
206
207
208
209
210
211
212
  "done": true,
  "total_duration": 14648695333,
  "load_duration": 3302671417,
  "prompt_eval_count": 14,
  "prompt_eval_duration": 286243000,
  "eval_count": 129,
  "eval_duration": 10931424000
}
```

213
214
215
#### Request (JSON mode)

```shell
216
curl http://localhost:11434/api/generate -d '{
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
  "model": "llama2",
  "prompt": "What color is the sky at different times of the day? Respond using JSON",
  "format": "json",
  "stream": false
}'
```

#### Response

```json
{
  "model": "llama2",
  "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,
  "total_duration": 4661289125,
  "load_duration": 1714434500,
  "prompt_eval_count": 36,
  "prompt_eval_duration": 264132000,
  "eval_count": 75,
  "eval_duration": 2112149000
}
```

The value of `response` will be a string containing JSON similar to:

```json
{
  "morning": {
    "color": "blue"
  },
  "noon": {
    "color": "blue-gray"
  },
  "afternoon": {
    "color": "warm gray"
  },
  "evening": {
    "color": "orange"
  }
}
```

#### Request (With options)
261
262
263
264

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.

```shell
265
curl http://localhost:11434/api/generate -d '{
266
  "model": "llama2",
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
  "prompt": "Why is the sky blue?",
  "stream": false,
  "options": {
    "num_keep": 5,
    "seed": 42,
    "num_predict": 100,
    "top_k": 20,
    "top_p": 0.9,
    "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,
288
    "num_ctx": 1024,
289
290
291
292
293
294
295
296
297
298
299
300
301
302
    "num_batch": 2,
    "num_gqa": 1,
    "num_gpu": 1,
    "main_gpu": 0,
    "low_vram": false,
    "f16_kv": true,
    "logits_all": false,
    "vocab_only": false,
    "use_mmap": true,
    "use_mlock": false,
    "embedding_only": false,
    "rope_frequency_base": 1.1,
    "rope_frequency_scale": 0.8,
    "num_thread": 8
303
  }
304
305
306
307
308
309
310
}'
```

#### Response

```json
{
311
  "model": "llama2",
312
313
314
315
316
317
318
319
320
321
322
323
324
325
  "created_at": "2023-08-04T19:22:45.499127Z",
  "response": "The sky is blue because it is the color of the sky.",
  "done": true,
  "total_duration": 5589157167,
  "load_duration": 3013701500,
  "sample_count": 114,
  "sample_duration": 81442000,
  "prompt_eval_count": 46,
  "prompt_eval_duration": 1160282000,
  "eval_count": 13,
  "eval_duration": 1325948000
}
```

326
## Generate a chat completion
327

Bruce MacDonald's avatar
Bruce MacDonald committed
328
329
330
331
332
333
334
335
336
337
338
```shell
POST /api/chat
```

Generate the next message in a chat with a provided model. This is a streaming endpoint, so there will be a series of responses. The final response object will include statistics and additional data from the request.

### Parameters

- `model`: (required) the [model name](#model-names)
- `messages`: the messages of the chat, this can be used to keep a chat memory

339
340
341
342
343
The `message` object has the following fields:

- `role`: the role of the message, either `system`, `user` or `assistant`
- `content`: the content of the message
- `images` (optional): a list of images to include in the message (for multimodal models such as `llava`)
344

Bruce MacDonald's avatar
Bruce MacDonald committed
345
346
347
348
349
350
351
352
353
354
Advanced parameters (optional):

- `format`: the format to return a response in. Currently the only accepted value is `json`
- `options`: additional model parameters listed in the documentation for the [Modelfile](./modelfile.md#valid-parameters-and-values) such as `temperature`
- `template`: the full prompt or prompt template (overrides what is defined in the `Modelfile`)
- `stream`: if `false` the response will be returned as a single response object, rather than a stream of objects

### Examples

#### Request
355

Bruce MacDonald's avatar
Bruce MacDonald committed
356
357
358
Send a chat message with a streaming response.

```shell
359
curl http://localhost:11434/api/chat -d '{
Bruce MacDonald's avatar
Bruce MacDonald committed
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
  "model": "llama2",
  "messages": [
    {
      "role": "user",
      "content": "why is the sky blue?"
    }
  ]
}'
```

#### Response

A stream of JSON objects is returned:

```json
{
  "model": "llama2",
  "created_at": "2023-08-04T08:52:19.385406455-07:00",
  "message": {
    "role": "assisant",
    "content": "The"
  },
  "done": false
}
```

Final response:

```json
{
  "model": "llama2",
  "created_at": "2023-08-04T19:22:45.499127Z",
  "done": true,
  "total_duration": 5589157167,
  "load_duration": 3013701500,
  "sample_count": 114,
  "sample_duration": 81442000,
  "prompt_eval_count": 46,
  "prompt_eval_duration": 1160282000,
  "eval_count": 113,
  "eval_duration": 1325948000
}
```

#### Request (With History)
405

Bruce MacDonald's avatar
Bruce MacDonald committed
406
407
408
Send a chat message with a conversation history.

```shell
409
curl http://localhost:11434/api/chat -d '{
Bruce MacDonald's avatar
Bruce MacDonald committed
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
  "model": "llama2",
  "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

A stream of JSON objects is returned:

```json
{
  "model": "llama2",
  "created_at": "2023-08-04T08:52:19.385406455-07:00",
  "message": {
    "role": "assisant",
    "content": "The"
  },
  "done": false
}
```

Final response:

```json
{
  "model": "llama2",
  "created_at": "2023-08-04T19:22:45.499127Z",
  "done": true,
  "total_duration": 5589157167,
  "load_duration": 3013701500,
  "sample_count": 114,
  "sample_duration": 81442000,
  "prompt_eval_count": 46,
  "prompt_eval_duration": 1160282000,
  "eval_count": 113,
  "eval_duration": 1325948000
}
```

462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
#### Request (with images)

Send a chat message with a conversation history.

```shell
curl http://localhost:11434/api/chat -d '{
  "model": "llama2",
  "messages": [
    {
      "role": "user",
      "content": "what is in this image?",
      "images": ["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"]
    },
  ]
}'
```

479
## Create a Model
Matt Williams's avatar
Matt Williams committed
480

Matt Williams's avatar
Matt Williams committed
481
```shell
482
483
484
POST /api/create
```

Michael Yang's avatar
Michael Yang committed
485
Create a model from a [`Modelfile`](./modelfile.md). It is recommended to set `modelfile` to the content of the Modelfile rather than just set `path`. This is a requirement for remote create. Remote model creation should also create any file blobs, fields such as `FROM` and `ADAPTER`, explicitly with the server using [Create a Blob](#create-a-blob) and the value to the path indicated in the response.
486

487
### Parameters
Matt Williams's avatar
Matt Williams committed
488

489
- `name`: name of the model to create
Jeffrey Morgan's avatar
Jeffrey Morgan committed
490
- `modelfile` (optional): contents of the Modelfile
491
- `stream`: (optional) if `false` the response will be returned as a single response object, rather than a stream of objects
Jeffrey Morgan's avatar
Jeffrey Morgan committed
492
- `path` (optional): path to the Modelfile
Matt Williams's avatar
Matt Williams committed
493

494
495
496
### Examples

#### Request
Matt Williams's avatar
Matt Williams committed
497

Matt Williams's avatar
Matt Williams committed
498
```shell
499
curl http://localhost:11434/api/create -d '{
500
  "name": "mario",
501
  "modelfile": "FROM llama2\nSYSTEM You are mario from Super Mario Bros."
502
}'
Matt Williams's avatar
Matt Williams committed
503
504
```

505
#### Response
Matt Williams's avatar
Matt Williams committed
506

Matt Williams's avatar
Matt Williams committed
507
A stream of JSON objects. When finished, `status` is `success`.
Matt Williams's avatar
Matt Williams committed
508

509
```json
Matt Williams's avatar
Matt Williams committed
510
511
512
513
514
{
  "status": "parsing modelfile"
}
```

Michael Yang's avatar
Michael Yang committed
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
### Check if a Blob Exists

```shell
HEAD /api/blobs/:digest
```

Check if a blob is known to the server.

#### Query Parameters

- `digest`: the SHA256 digest of the blob

#### Examples

##### Request

```shell
curl -I http://localhost:11434/api/blobs/sha256:29fdb92e57cf0827ded04ae6461b5931d01fa595843f55d36f5b275a52087dd2
```

##### Response

Return 200 OK if the blob exists, 404 Not Found if it does not.

Michael Yang's avatar
Michael Yang committed
539
### Create a Blob
Michael Yang's avatar
Michael Yang committed
540
541
542
543
544
545
546

```shell
POST /api/blobs/:digest
```

Create a blob from a file. Returns the server file path.

Michael Yang's avatar
Michael Yang committed
547
#### Query Parameters
Michael Yang's avatar
Michael Yang committed
548
549
550

- `digest`: the expected SHA256 digest of the file

Michael Yang's avatar
Michael Yang committed
551
#### Examples
Michael Yang's avatar
Michael Yang committed
552

Michael Yang's avatar
Michael Yang committed
553
554
##### Request

Michael Yang's avatar
Michael Yang committed
555
```shell
Michael Yang's avatar
Michael Yang committed
556
curl -T model.bin -X POST http://localhost:11434/api/blobs/sha256:29fdb92e57cf0827ded04ae6461b5931d01fa595843f55d36f5b275a52087dd2
Michael Yang's avatar
Michael Yang committed
557
558
```

Michael Yang's avatar
Michael Yang committed
559
##### Response
Michael Yang's avatar
Michael Yang committed
560

Michael Yang's avatar
Michael Yang committed
561
Return 201 Created if the blob was successfully created.
Michael Yang's avatar
Michael Yang committed
562

563
## List Local Models
Matt Williams's avatar
Matt Williams committed
564

Matt Williams's avatar
Matt Williams committed
565
```shell
566
GET /api/tags
Matt Williams's avatar
Matt Williams committed
567
568
```

569
List models that are available locally.
Matt Williams's avatar
Matt Williams committed
570

571
572
573
### Examples

#### Request
Matt Williams's avatar
Matt Williams committed
574

Matt Williams's avatar
Matt Williams committed
575
```shell
576
577
curl http://localhost:11434/api/tags
```
Matt Williams's avatar
Matt Williams committed
578

579
#### Response
Matt Williams's avatar
Matt Williams committed
580

581
582
A single JSON object will be returned.

583
```json
Matt Williams's avatar
Matt Williams committed
584
585
586
{
  "models": [
    {
587
      "name": "llama2",
588
589
590
591
592
593
      "modified_at": "2023-08-02T17:02:23.713454393-07:00",
      "size": 3791730596
    },
    {
      "name": "llama2:13b",
      "modified_at": "2023-08-08T12:08:38.093596297-07:00",
Matt Williams's avatar
Matt Williams committed
594
      "size": 7323310500
Matt Williams's avatar
Matt Williams committed
595
596
    }
  ]
Matt Williams's avatar
Matt Williams committed
597
598
599
}
```

Matt Williams's avatar
Matt Williams committed
600
601
602
603
604
605
## Show Model Information

```shell
POST /api/show
```

606
Show information about a model including details, modelfile, template, parameters, license, and system prompt.
Matt Williams's avatar
Matt Williams committed
607
608
609
610

### Parameters

- `name`: name of the model to show
Matt Williams's avatar
Matt Williams committed
611

612
613
614
### Examples

#### Request
Matt Williams's avatar
Matt Williams committed
615

616
```shell
Matt Williams's avatar
Matt Williams committed
617
curl http://localhost:11434/api/show -d '{
618
  "name": "llama2"
Matt Williams's avatar
Matt Williams committed
619
}'
Matt Williams's avatar
Matt Williams committed
620
```
Matt Williams's avatar
Matt Williams committed
621

622
#### Response
Matt Williams's avatar
Matt Williams committed
623
624
625

```json
{
626
627
628
629
630
631
632
633
634
635
  "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 mike/llava:latest\nTEMPLATE \"\"\"\nUSER:{{ .Prompt }}\nASSISTANT:\n\"\"\"\nPARAMETER num_ctx 4096",
  "parameters": "num_ctx                        4096",
  "template": "\nUSER:{{ .Prompt }}\nASSISTANT:\n",
  "license:": "<license>",
  "details": {
    "format": "gguf",
    "families": ["llama", "clip"],
    "parameter_size": "7B",
    "quantization_level": "Q4_0"
  }
Matt Williams's avatar
Matt Williams committed
636
637
638
639
640
641
}
```

## Copy a Model

```shell
642
POST /api/copy
Matt Williams's avatar
Matt Williams committed
643
```
644

645
Copy a model. Creates a model with another name from an existing model.
Matt Williams's avatar
Matt Williams committed
646

647
648
649
### Examples

#### Request
Matt Williams's avatar
Matt Williams committed
650

Matt Williams's avatar
Matt Williams committed
651
```shell
652
curl http://localhost:11434/api/copy -d '{
653
  "source": "llama2",
654
  "destination": "llama2-backup"
Matt Williams's avatar
Matt Williams committed
655
656
657
}'
```

658
#### Response
659
660
661

The only response is a 200 OK if successful.

Matt Williams's avatar
Matt Williams committed
662
## Delete a Model
Matt Williams's avatar
Matt Williams committed
663

Matt Williams's avatar
Matt Williams committed
664
```shell
665
DELETE /api/delete
Matt Williams's avatar
Matt Williams committed
666
667
```

668
Delete a model and its data.
Matt Williams's avatar
Matt Williams committed
669

670
### Parameters
Matt Williams's avatar
Matt Williams committed
671

672
- `name`: model name to delete
Matt Williams's avatar
Matt Williams committed
673

674
675
676
### Examples

#### Request
Matt Williams's avatar
Matt Williams committed
677

Matt Williams's avatar
Matt Williams committed
678
```shell
679
680
curl -X DELETE http://localhost:11434/api/delete -d '{
  "name": "llama2:13b"
Matt Williams's avatar
Matt Williams committed
681
682
683
}'
```

684
#### Response
685
686
687

If successful, the only response is a 200 OK.

688
## Pull a Model
Matt Williams's avatar
Matt Williams committed
689

Matt Williams's avatar
Matt Williams committed
690
```shell
691
692
693
POST /api/pull
```

Matt Williams's avatar
Matt Williams committed
694
Download a model from the ollama library. Cancelled pulls are resumed from where they left off, and multiple calls will share the same download progress.
Matt Williams's avatar
Matt Williams committed
695

696
### Parameters
Matt Williams's avatar
Matt Williams committed
697

698
- `name`: name of the model to pull
Matt Williams's avatar
Matt Williams committed
699
- `insecure`: (optional) allow insecure connections to the library. Only use this if you are pulling from your own library during development.
700
- `stream`: (optional) if `false` the response will be returned as a single response object, rather than a stream of objects
Matt Williams's avatar
Matt Williams committed
701

702
703
704
### Examples

#### Request
Matt Williams's avatar
Matt Williams committed
705

Matt Williams's avatar
Matt Williams committed
706
```shell
707
curl http://localhost:11434/api/pull -d '{
708
  "name": "llama2"
709
}'
Matt Williams's avatar
Matt Williams committed
710
711
```

712
#### Response
713

714
715
716
717
718
719
720
721
722
723
724
725
If `stream` is not specified, or set to `true`, a stream of JSON objects is returned:

The first object is the manifest:

```json
{
  "status": "pulling manifest"
}
```

Then there is a series of downloading responses. Until any of the download is completed, the `completed` key may not be included. The number of files to be downloaded depends on the number of layers specified in the manifest.

726
```json
Matt Williams's avatar
Matt Williams committed
727
{
728
729
  "status": "downloading digestname",
  "digest": "digestname",
730
  "total": 2142590208,
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
  "completed": 241970
}
```

After all the files are downloaded, the final responses are:

```json
{
    "status": "verifying sha256 digest"
}
{
    "status": "writing manifest"
}
{
    "status": "removing any unused layers"
}
{
    "status": "success"
}
```

if `stream` is set to false, then the response is a single JSON object:

```json
{
  "status": "success"
Matt Williams's avatar
Matt Williams committed
757
758
}
```
759

Matt Williams's avatar
Matt Williams committed
760
761
762
763
764
765
766
767
768
769
770
## Push a Model

```shell
POST /api/push
```

Upload a model to a model library. Requires registering for ollama.ai and adding a public key first.

### Parameters

- `name`: name of the model to push in the form of `<namespace>/<model>:<tag>`
771
- `insecure`: (optional) allow insecure connections to the library. Only use this if you are pushing to your library during development.
772
- `stream`: (optional) if `false` the response will be returned as a single response object, rather than a stream of objects
Matt Williams's avatar
Matt Williams committed
773

774
775
776
### Examples

#### Request
Matt Williams's avatar
Matt Williams committed
777
778

```shell
779
curl http://localhost:11434/api/push -d '{
Matt Williams's avatar
Matt Williams committed
780
781
782
783
  "name": "mattw/pygmalion:latest"
}'
```

784
#### Response
785

786
If `stream` is not specified, or set to `true`, a stream of JSON objects is returned:
Matt Williams's avatar
Matt Williams committed
787
788

```json
789
{ "status": "retrieving manifest" }
790
```
Matt Williams's avatar
Matt Williams committed
791
792
793
794
795

and then:

```json
{
796
797
798
  "status": "starting upload",
  "digest": "sha256:bc07c81de745696fdf5afca05e065818a8149fb0c77266fb584d9b2cba3711ab",
  "total": 1928429856
Matt Williams's avatar
Matt Williams committed
799
800
801
802
803
804
805
}
```

Then there is a series of uploading responses:

```json
{
806
807
808
809
  "status": "starting upload",
  "digest": "sha256:bc07c81de745696fdf5afca05e065818a8149fb0c77266fb584d9b2cba3711ab",
  "total": 1928429856
}
Matt Williams's avatar
Matt Williams committed
810
811
812
813
814
815
816
817
818
```

Finally, when the upload is complete:

```json
{"status":"pushing manifest"}
{"status":"success"}
```

819
820
821
If `stream` is set to `false`, then the response is a single JSON object:

```json
822
{ "status": "success" }
823
824
```

Matt Williams's avatar
Matt Williams committed
825
826
827
## Generate Embeddings

```shell
828
829
830
831
832
833
834
835
836
837
POST /api/embeddings
```

Generate embeddings from a model

### Parameters

- `model`: name of model to generate embeddings from
- `prompt`: text to generate embeddings for

838
839
840
841
Advanced parameters:

- `options`: additional model parameters listed in the documentation for the [Modelfile](./modelfile.md#valid-parameters-and-values) such as `temperature`

842
843
844
### Examples

#### Request
845

Matt Williams's avatar
Matt Williams committed
846
```shell
847
curl http://localhost:11434/api/embeddings -d '{
848
  "model": "llama2",
849
850
851
852
  "prompt": "Here is an article about llamas..."
}'
```

853
#### Response
854
855
856

```json
{
Alexander F. Rødseth's avatar
Alexander F. Rødseth committed
857
  "embedding": [
858
859
860
    0.5670403838157654, 0.009260174818336964, 0.23178744316101074, -0.2916173040866852, -0.8924556970596313,
    0.8785552978515625, -0.34576427936553955, 0.5742510557174683, -0.04222835972905159, -0.137906014919281
  ]
Costa Alexoglou's avatar
Costa Alexoglou committed
861
862
}
```