File size: 48,366 Bytes
bf6d237
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
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
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
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
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
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
405
406
407
408
409
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
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
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
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
{
  "openapi": "3.1.0",
  "info": {
    "title": "PrivateGPT",
    "summary": "PrivateGPT is a production-ready AI project that allows you to ask questions to your documents using the power of Large Language Models (LLMs), even in scenarios without Internet connection. 100% private, no data leaves your execution environment at any point.",
    "description": "## Introduction\n\nPrivateGPT provides an **API** containing all the building blocks required to build\n**private, context-aware AI applications**. The API follows and extends OpenAI API standard, and supports\nboth normal and streaming responses.\n\nThe API is divided in two logical blocks:\n\n- High-level API, abstracting all the complexity of a RAG (Retrieval Augmented Generation) pipeline implementation:\n    - Ingestion of documents: internally managing document parsing, splitting, metadata extraction,\n      embedding generation and storage.\n    - Chat & Completions using context from ingested documents: abstracting the retrieval of context, the prompt\n      engineering and the response generation.\n- Low-level API, allowing advanced users to implement their own complex pipelines:\n    - Embeddings generation: based on a piece of text.\n    - Contextual chunks retrieval: given a query, returns the most relevant chunks of text from the ingested\n      documents.\n\n> A working **Gradio UI client** is provided to test the API, together with a set of\n> useful tools such as bulk model download script, ingestion script, documents folder\n> watch, etc.\n\n## Quick Local Installation steps\n\nThe steps in `Installation and Settings` section are better explained and cover more\nsetup scenarios. But if you are looking for a quick setup guide, here it is:\n\n```\n# Clone the repo\ngit clone https://github.com/imartinez/privateGPT\ncd privateGPT\n\n# Install Python 3.11\npyenv install 3.11\npyenv local 3.11\n\n# Install dependencies\npoetry install --with ui,local\n\n# Download Embedding and LLM models\npoetry run python scripts/setup\n\n# (Optional) For Mac with Metal GPU, enable it. Check Installation and Settings section \nto know how to enable GPU on other platforms\nCMAKE_ARGS=\"-DLLAMA_METAL=on\" pip install --force-reinstall --no-cache-dir llama-cpp-python\n\n# Run the local server  \nPGPT_PROFILES=local make run\n\n# Note: on Mac with Metal you should see a ggml_metal_add_buffer log, stating GPU is \nbeing used\n\n# Navigate to the UI and try it out! \nhttp://localhost:8001/\n```\n\n## Installation and Settings\n\n### Base requirements to run PrivateGPT\n\n* Git clone PrivateGPT repository, and navigate to it:\n\n```\n  git clone https://github.com/imartinez/privateGPT\n  cd privateGPT\n```\n\n* Install Python 3.11. Ideally through a python version manager like `pyenv`.\n  Python 3.12\n  should work too. Earlier python versions are not supported.\n    * osx/linux: [pyenv](https://github.com/pyenv/pyenv)\n    * windows: [pyenv-win](https://github.com/pyenv-win/pyenv-win)\n\n```  \npyenv install 3.11\npyenv local 3.11\n```\n\n* Install [Poetry](https://python-poetry.org/docs/#installing-with-the-official-installer) for dependency management:\n\n* Have a valid C++ compiler like gcc. See [Troubleshooting: C++ Compiler](#troubleshooting-c-compiler) for more details.\n\n* Install `make` for scripts:\n    * osx: (Using homebrew): `brew install make`\n    * windows: (Using chocolatey) `choco install make`\n\n### Install dependencies\n\nInstall the dependencies:\n\n```bash\npoetry install --with ui\n```\n\nVerify everything is working by running `make run` (or `poetry run python -m private_gpt`) and navigate to\nhttp://localhost:8001. You should see a [Gradio UI](https://gradio.app/) **configured with a mock LLM** that will\necho back the input. Later we'll see how to configure a real LLM.\n\n### Settings\n\n> Note: the default settings of PrivateGPT work out-of-the-box for a 100% local setup. Skip this section if you just\n> want to test PrivateGPT locally, and come back later to learn about more configuration options.\n\nPrivateGPT is configured through *profiles* that are defined using yaml files, and selected through env variables.\nThe full list of properties configurable can be found in `settings.yaml`\n\n#### env var `PGPT_SETTINGS_FOLDER`\n\nThe location of the settings folder. Defaults to the root of the project.\nShould contain the default `settings.yaml` and any other `settings-{profile}.yaml`.\n\n#### env var `PGPT_PROFILES`\n\nBy default, the profile definition in `settings.yaml` is loaded.\nUsing this env var you can load additional profiles; format is a comma separated list of profile names.\nThis will merge `settings-{profile}.yaml` on top of the base settings file.\n\nFor example:\n`PGPT_PROFILES=local,cuda` will load `settings-local.yaml`\nand `settings-cuda.yaml`, their contents will be merged with\nlater profiles properties overriding values of earlier ones like `settings.yaml`.\n\nDuring testing, the `test` profile will be active along with the default, therefore `settings-test.yaml`\nfile is required.\n\n#### Environment variables expansion\n\nConfiguration files can contain environment variables,\nthey will be expanded at runtime.\n\nExpansion must follow the pattern `${VARIABLE_NAME:default_value}`.\n\nFor example, the following configuration will use the value of the `PORT`\nenvironment variable or `8001` if it's not set.\nMissing variables with no default will produce an error.\n\n```yaml\nserver:\n  port: ${PORT:8001}\n```\n\n### Local LLM requirements\n\nInstall extra dependencies for local execution:\n\n```bash\npoetry install --with local\n```\n\nFor PrivateGPT to run fully locally GPU acceleration is required\n(CPU execution is possible, but very slow), however,\ntypical Macbook laptops or window desktops with mid-range GPUs lack VRAM to run\neven the smallest LLMs. For that reason\n**local execution is only supported for models compatible with [llama.cpp](https://github.com/ggerganov/llama.cpp)**\n\nThese two models are known to work well:\n\n* https://huggingface.co/TheBloke/Llama-2-7B-chat-GGUF\n* https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.1-GGUF (recommended)\n\nTo ease the installation process, use the `setup` script that will download both\nthe embedding and the LLM model and place them in the correct location (under `models` folder):\n\n```bash\npoetry run python scripts/setup\n```\n\nIf you are ok with CPU execution, you can skip the rest of this section.\n\nAs stated before, llama.cpp is required and in\nparticular [llama-cpp-python](https://github.com/abetlen/llama-cpp-python)\nis used.\n\n> It's highly encouraged that you fully read llama-cpp and llama-cpp-python documentation relevant to your platform.\n> Running into installation issues is very likely, and you'll need to troubleshoot them yourself.\n\n#### Customizing low level parameters\n\nCurrently not all the parameters of llama-cpp and llama-cpp-python are available at PrivateGPT's `settings.yaml` file. In case you need to customize parameters such as the number of layers loaded into the GPU, you might change these at the `llm_component.py` file under the `private_gpt/components/llm/llm_component.py`. If you are getting an out of memory error, you might also try a smaller model or stick to the proposed recommended models, instead of custom tuning the parameters.\n\n#### OSX GPU support\n\nYou will need to build [llama.cpp](https://github.com/ggerganov/llama.cpp) with\nmetal support. To do that run:\n\n```bash\nCMAKE_ARGS=\"-DLLAMA_METAL=on\" pip install --force-reinstall --no-cache-dir llama-cpp-python\n```\n\n#### Windows NVIDIA GPU support\n\nWindows GPU support is done through CUDA.\nFollow the instructions on the original [llama.cpp](https://github.com/ggerganov/llama.cpp) repo to install the required\ndependencies.\n\nSome tips to get it working with an NVIDIA card and CUDA (Tested on Windows 10 with CUDA 11.5 RTX 3070):\n\n* Install latest VS2022 (and build tools) https://visualstudio.microsoft.com/vs/community/\n* Install CUDA toolkit https://developer.nvidia.com/cuda-downloads\n* Verify your installation is correct by running `nvcc --version` and `nvidia-smi`, ensure your CUDA version is up to\n  date and your GPU is detected.\n* [Optional] Install CMake to troubleshoot building issues by compiling llama.cpp directly https://cmake.org/download/\n\nIf you have all required dependencies properly configured running the\nfollowing powershell command should succeed.\n\n```powershell\n$env:CMAKE_ARGS='-DLLAMA_CUBLAS=on'; poetry run pip install --force-reinstall --no-cache-dir llama-cpp-python\n```\n\nIf your installation was correct, you should see a message similar to the following next\ntime you start the server `BLAS = 1`.\n\n```\nllama_new_context_with_model: total VRAM used: 4857.93 MB (model: 4095.05 MB, context: 762.87 MB)\nAVX = 1 | AVX2 = 1 | AVX512 = 0 | AVX512_VBMI = 0 | AVX512_VNNI = 0 | FMA = 1 | NEON = 0 | ARM_FMA = 0 | F16C = 1 | FP16_VA = 0 | WASM_SIMD = 0 | BLAS = 1 | SSE3 = 1 | SSSE3 = 0 | VSX = 0 | \n```\n\nNote that llama.cpp offloads matrix calculations to the GPU but the performance is\nstill hit heavily due to latency between CPU and GPU communication. You might need to tweak\nbatch sizes and other parameters to get the best performance for your particular system.\n\n#### Linux NVIDIA GPU support and Windows-WSL\n\nLinux GPU support is done through CUDA.\nFollow the instructions on the original [llama.cpp](https://github.com/ggerganov/llama.cpp) repo to install the required\nexternal\ndependencies.\n\nSome tips:\n\n* Make sure you have an up-to-date C++ compiler\n* Install CUDA toolkit https://developer.nvidia.com/cuda-downloads\n* Verify your installation is correct by running `nvcc --version` and `nvidia-smi`, ensure your CUDA version is up to\n  date and your GPU is detected.\n\nAfter that running the following command in the repository will install llama.cpp with GPU support:\n\n`\nCMAKE_ARGS='-DLLAMA_CUBLAS=on' poetry run pip install --force-reinstall --no-cache-dir llama-cpp-python\n`\n\nIf your installation was correct, you should see a message similar to the following next\ntime you start the server `BLAS = 1`.\n\n```\nllama_new_context_with_model: total VRAM used: 4857.93 MB (model: 4095.05 MB, context: 762.87 MB)\nAVX = 1 | AVX2 = 1 | AVX512 = 0 | AVX512_VBMI = 0 | AVX512_VNNI = 0 | FMA = 1 | NEON = 0 | ARM_FMA = 0 | F16C = 1 | FP16_VA = 0 | WASM_SIMD = 0 | BLAS = 1 | SSE3 = 1 | SSSE3 = 0 | VSX = 0 | \n```\n\n#### Vectorstores\nPrivateGPT supports [Chroma](https://www.trychroma.com/), [Qdrant](https://qdrant.tech/) as vectorstore providers. Chroma being the default.\n\nTo enable Qdrant, set the `vectorstore.database` property in the `settings.yaml` file to `qdrant` and install the `qdrant` extra.\n\n```bash\npoetry install --extras qdrant\n```\n\nBy default Qdrant tries to connect to an instance at `http://localhost:3000`.\n\nQdrant settings can be configured by setting values to the `qdrant` propery in the `settings.yaml` file.\n\nThe available configuration options are:\n| Field        | Description |\n|--------------|-------------|\n| location     | If `:memory:` - use in-memory Qdrant instance.<br>If `str` - use it as a `url` parameter.|\n| url          | Either host or str of 'Optional[scheme], host, Optional[port], Optional[prefix]'.<br> Eg. `http://localhost:6333` |\n| port         | Port of the REST API interface. Default: `6333` |\n| grpc_port    | Port of the gRPC interface. Default: `6334` |\n| prefer_grpc  | If `true` - use gRPC interface whenever possible in custom methods. |\n| https        | If `true` - use HTTPS(SSL) protocol.|\n| api_key      | API key for authentication in Qdrant Cloud.|\n| prefix       | If set, add `prefix` to the REST URL path.<br>Example: `service/v1` will result in `http://localhost:6333/service/v1/{qdrant-endpoint}` for REST API.|\n| timeout      | Timeout for REST and gRPC API requests.<br>Default: 5.0 seconds for REST and unlimited for gRPC |\n| host         | Host name of Qdrant service. If url and host are not set, defaults to 'localhost'.|\n| path         | Persistence path for QdrantLocal. Eg. `local_data/private_gpt/qdrant`|\n| force_disable_check_same_thread         | Force disable check_same_thread for QdrantLocal sqlite connection.|\n\n#### Known issues and Troubleshooting\n\nExecution of LLMs locally still has a lot of sharp edges, specially when running on non Linux platforms.\nYou might encounter several issues:\n\n* Performance: RAM or VRAM usage is very high, your computer might experience slowdowns or even crashes.\n* GPU Virtualization on Windows and OSX: Simply not possible with docker desktop, you have to run the server directly on\n  the host.\n* Building errors: Some of PrivateGPT dependencies need to build native code, and they might fail on some platforms.\n  Most likely you are missing some dev tools in your machine (updated C++ compiler, CUDA is not on PATH, etc.).\n  If you encounter any of these issues, please open an issue and we'll try to help.\n\n#### Troubleshooting: C++ Compiler\n\nIf you encounter an error while building a wheel during the `pip install` process, you may need to install a C++\ncompiler on your computer.\n\n**For Windows 10/11**\n\nTo install a C++ compiler on Windows 10/11, follow these steps:\n\n1. Install Visual Studio 2022.\n2. Make sure the following components are selected:\n    * Universal Windows Platform development\n    * C++ CMake tools for Windows\n3. Download the MinGW installer from the [MinGW website](https://sourceforge.net/projects/mingw/).\n4. Run the installer and select the `gcc` component.\n\n** For OSX **\n\n1. Check if you have a C++ compiler installed, Xcode might have done it for you. for example running `gcc`.\n2. If not, you can install clang or gcc with homebrew `brew install gcc`\n\n#### Troubleshooting: Mac Running Intel\n\nWhen running a Mac with Intel hardware (not M1), you may run into _clang: error: the clang compiler does not support '\n-march=native'_ during pip install.\n\nIf so set your archflags during pip install. eg: _ARCHFLAGS=\"-arch x86_64\" pip3 install -r requirements.txt_\n\n## Running the Server\n\nAfter following the installation steps you should be ready to go. Here are some common run setups:\n\n### Running 100% locally\n\nMake sure you have followed the *Local LLM requirements* section before moving on.\n\nThis command will start PrivateGPT using the `settings.yaml` (default profile) together with the `settings-local.yaml`\nconfiguration files. By default, it will enable both the API and the Gradio UI. Run:\n\n```\nPGPT_PROFILES=local make run\n``` \n\nor\n\n```\nPGPT_PROFILES=local poetry run python -m private_gpt\n```\n\nWhen the server is started it will print a log *Application startup complete*.\nNavigate to http://localhost:8001 to use the Gradio UI or to http://localhost:8001/docs (API section) to try the API\nusing Swagger UI.\n\n### Local server using OpenAI as LLM\n\nIf you cannot run a local model (because you don't have a GPU, for example) or for testing purposes, you may\ndecide to run PrivateGPT using OpenAI as the LLM.\n\nIn order to do so, create a profile `settings-openai.yaml` with the following contents:\n\n```yaml\nllm:\n  mode: openai\n\nopenai:\n  api_key: <your_openai_api_key>  # You could skip this configuration and use the OPENAI_API_KEY env var instead\n```\n\nAnd run PrivateGPT loading that profile you just created:\n\n```PGPT_PROFILES=openai make run```\n\nor\n\n```PGPT_PROFILES=openai poetry run python -m private_gpt```\n\n> Note this will still use the local Embeddings model, as it is ok to use it on a CPU.\n> We'll support using OpenAI embeddings in a future release.\n\nWhen the server is started it will print a log *Application startup complete*.\nNavigate to http://localhost:8001 to use the Gradio UI or to http://localhost:8001/docs (API section) to try the API.\nYou'll notice the speed and quality of response is higher, given you are using OpenAI's servers for the heavy\ncomputations.\n\n### Use AWS's Sagemaker\n\n\ud83d\udea7 Under construction \ud83d\udea7\n\n## Gradio UI user manual\n\nGradio UI is a ready to use way of testing most of PrivateGPT API functionalities.\n\n![Gradio PrivateGPT](https://lh3.googleusercontent.com/drive-viewer/AK7aPaD_Hc-A8A9ooMe-hPgm_eImgsbxAjb__8nFYj8b_WwzvL1Gy90oAnp1DfhPaN6yGiEHCOXs0r77W1bYHtPzlVwbV7fMsA=s1600)\n\n### Execution Modes\n\nIt has 3 modes of execution (you can select in the top-left):\n\n* Query Docs: uses the context from the\n  ingested documents to answer the questions posted in the chat. It also takes\n  into account previous chat messages as context.\n    * Makes use of `/chat/completions` API with `use_context=true` and no\n      `context_filter`.\n* Search in Docs: fast search that returns the 4 most related text\n  chunks, together with their source document and page.\n    * Makes use of `/chunks` API with no `context_filter`, `limit=4` and\n      `prev_next_chunks=0`.\n* LLM Chat: simple, non-contextual chat with the LLM. The ingested documents won't\n  be taken into account, only the previous messages.\n    * Makes use of `/chat/completions` API with `use_context=false`.\n\n### Document Ingestion\n\nIngest documents by using the `Upload a File` button. You can check the progress of\nthe ingestion in the console logs of the server.\n\nThe list of ingested files is shown below the button.\n\nIf you want to delete the ingested documents, refer to *Reset Local documents\ndatabase* section in the documentation.\n\n### Chat\n\nNormal chat interface, self-explanatory ;)\n\nYou can check the actual prompt being passed to the LLM by looking at the logs of\nthe server. We'll add better observability in future releases.\n\n## Deployment options\n\n\ud83d\udea7 We are working on Dockerized deployment guidelines \ud83d\udea7\n\n## Observability\n\nBasic logs are enabled using LlamaIndex\nbasic logging (for example ingestion progress or LLM prompts and answers).\n\n\ud83d\udea7 We are working on improved Observability. \ud83d\udea7\n\n## Ingesting & Managing Documents\n\n\ud83d\udea7 Document Update and Delete are still WIP. \ud83d\udea7\n\nThe ingestion of documents can be done in different ways:\n\n* Using the `/ingest` API\n* Using the Gradio UI\n* Using the Bulk Local Ingestion functionality (check next section)\n\n### Bulk Local Ingestion\n\nWhen you are running PrivateGPT in a fully local setup, you can ingest a complete folder for convenience (containing\npdf, text files, etc.)\nand optionally watch changes on it with the command:\n\n```bash\nmake ingest /path/to/folder -- --watch\n```\n\nTo log the processed and failed files to an additional file, use:\n\n```bash\nmake ingest /path/to/folder -- --watch --log-file /path/to/log/file.log\n```\n\nAfter ingestion is complete, you should be able to chat with your documents\nby navigating to http://localhost:8001 and using the option `Query documents`,\nor using the completions / chat API.\n\n### Reset Local documents database\n\nWhen running in a local setup, you can remove all ingested documents by simply\ndeleting all contents of `local_data` folder (except .gitignore).\n\n## API\n\nAs explained in the introduction, the API contains high level APIs (ingestion and chat/completions) and low level APIs\n(embeddings and chunk retrieval). In this section the different specific API calls are explained.\n",
    "contact": {
      "url": "https://github.com/imartinez/privateGPT"
    },
    "license": {
      "name": "Apache 2.0",
      "url": "https://www.apache.org/licenses/LICENSE-2.0.html"
    },
    "version": "0.1.0",
    "x-logo": {
      "url": "https://lh3.googleusercontent.com/drive-viewer/AK7aPaD_iNlMoTquOBsw4boh4tIYxyEuhz6EtEs8nzq3yNkNAK00xGjE1KUCmPJSk3TYOjcs6tReG6w_cLu1S7L_gPgT9z52iw=s2560"
    }
  },
  "paths": {
    "/v1/completions": {
      "post": {
        "tags": [
          "Contextual Completions"
        ],
        "summary": "Completion",
        "description": "We recommend most users use our Chat completions API.\n\nGiven a prompt, the model will return one predicted completion. If `use_context`\nis set to `true`, the model will use context coming from the ingested documents\nto create the response. The documents being used can be filtered using the\n`context_filter` and passing the document IDs to be used. Ingested documents IDs\ncan be found using `/ingest/list` endpoint. If you want all ingested documents to\nbe used, remove `context_filter` altogether.\n\nWhen using `'include_sources': true`, the API will return the source Chunks used\nto create the response, which come from the context provided.\n\nWhen using `'stream': true`, the API will return data chunks following [OpenAI's\nstreaming model](https://platform.openai.com/docs/api-reference/chat/streaming):\n```\n{\"id\":\"12345\",\"object\":\"completion.chunk\",\"created\":1694268190,\n\"model\":\"private-gpt\",\"choices\":[{\"index\":0,\"delta\":{\"content\":\"Hello\"},\n\"finish_reason\":null}]}\n```",
        "operationId": "prompt_completion_v1_completions_post",
        "requestBody": {
          "content": {
            "application/json": {
              "schema": {
                "$ref": "#/components/schemas/CompletionsBody"
              }
            }
          },
          "required": true
        },
        "responses": {
          "200": {
            "description": "Successful Response",
            "content": {
              "application/json": {
                "schema": {
                  "$ref": "#/components/schemas/OpenAICompletion"
                }
              }
            }
          },
          "422": {
            "description": "Validation Error",
            "content": {
              "application/json": {
                "schema": {
                  "$ref": "#/components/schemas/HTTPValidationError"
                }
              }
            }
          }
        }
      }
    },
    "/v1/chat/completions": {
      "post": {
        "tags": [
          "Contextual Completions"
        ],
        "summary": "Chat Completion",
        "description": "Given a list of messages comprising a conversation, return a response.\n\nIf `use_context` is set to `true`, the model will use context coming\nfrom the ingested documents to create the response. The documents being used can\nbe filtered using the `context_filter` and passing the document IDs to be used.\nIngested documents IDs can be found using `/ingest/list` endpoint. If you want\nall ingested documents to be used, remove `context_filter` altogether.\n\nWhen using `'include_sources': true`, the API will return the source Chunks used\nto create the response, which come from the context provided.\n\nWhen using `'stream': true`, the API will return data chunks following [OpenAI's\nstreaming model](https://platform.openai.com/docs/api-reference/chat/streaming):\n```\n{\"id\":\"12345\",\"object\":\"completion.chunk\",\"created\":1694268190,\n\"model\":\"private-gpt\",\"choices\":[{\"index\":0,\"delta\":{\"content\":\"Hello\"},\n\"finish_reason\":null}]}\n```",
        "operationId": "chat_completion_v1_chat_completions_post",
        "requestBody": {
          "content": {
            "application/json": {
              "schema": {
                "$ref": "#/components/schemas/ChatBody"
              }
            }
          },
          "required": true
        },
        "responses": {
          "200": {
            "description": "Successful Response",
            "content": {
              "application/json": {
                "schema": {
                  "$ref": "#/components/schemas/OpenAICompletion"
                }
              }
            }
          },
          "422": {
            "description": "Validation Error",
            "content": {
              "application/json": {
                "schema": {
                  "$ref": "#/components/schemas/HTTPValidationError"
                }
              }
            }
          }
        }
      }
    },
    "/v1/chunks": {
      "post": {
        "tags": [
          "Context Chunks"
        ],
        "summary": "Chunks Retrieval",
        "description": "Given a `text`, returns the most relevant chunks from the ingested documents.\n\nThe returned information can be used to generate prompts that can be\npassed to `/completions` or `/chat/completions` APIs. Note: it is usually a very\nfast API, because only the Embeddings model is involved, not the LLM. The\nreturned information contains the relevant chunk `text` together with the source\n`document` it is coming from. It also contains a score that can be used to\ncompare different results.\n\nThe max number of chunks to be returned is set using the `limit` param.\n\nPrevious and next chunks (pieces of text that appear right before or after in the\ndocument) can be fetched by using the `prev_next_chunks` field.\n\nThe documents being used can be filtered using the `context_filter` and passing\nthe document IDs to be used. Ingested documents IDs can be found using\n`/ingest/list` endpoint. If you want all ingested documents to be used,\nremove `context_filter` altogether.",
        "operationId": "chunks_retrieval_v1_chunks_post",
        "requestBody": {
          "content": {
            "application/json": {
              "schema": {
                "$ref": "#/components/schemas/ChunksBody"
              }
            }
          },
          "required": true
        },
        "responses": {
          "200": {
            "description": "Successful Response",
            "content": {
              "application/json": {
                "schema": {
                  "$ref": "#/components/schemas/ChunksResponse"
                }
              }
            }
          },
          "422": {
            "description": "Validation Error",
            "content": {
              "application/json": {
                "schema": {
                  "$ref": "#/components/schemas/HTTPValidationError"
                }
              }
            }
          }
        }
      }
    },
    "/v1/ingest": {
      "post": {
        "tags": [
          "Ingestion"
        ],
        "summary": "Ingest",
        "description": "Ingests and processes a file, storing its chunks to be used as context.\n\nThe context obtained from files is later used in\n`/chat/completions`, `/completions`, and `/chunks` APIs.\n\nMost common document\nformats are supported, but you may be prompted to install an extra dependency to\nmanage a specific file type.\n\nA file can generate different Documents (for example a PDF generates one Document\nper page). All Documents IDs are returned in the response, together with the\nextracted Metadata (which is later used to improve context retrieval). Those IDs\ncan be used to filter the context used to create responses in\n`/chat/completions`, `/completions`, and `/chunks` APIs.",
        "operationId": "ingest_v1_ingest_post",
        "requestBody": {
          "content": {
            "multipart/form-data": {
              "schema": {
                "$ref": "#/components/schemas/Body_ingest_v1_ingest_post"
              }
            }
          },
          "required": true
        },
        "responses": {
          "200": {
            "description": "Successful Response",
            "content": {
              "application/json": {
                "schema": {
                  "$ref": "#/components/schemas/IngestResponse"
                }
              }
            }
          },
          "422": {
            "description": "Validation Error",
            "content": {
              "application/json": {
                "schema": {
                  "$ref": "#/components/schemas/HTTPValidationError"
                }
              }
            }
          }
        }
      }
    },
    "/v1/ingest/list": {
      "get": {
        "tags": [
          "Ingestion"
        ],
        "summary": "List Ingested",
        "description": "Lists already ingested Documents including their Document ID and metadata.\n\nThose IDs can be used to filter the context used to create responses\nin `/chat/completions`, `/completions`, and `/chunks` APIs.",
        "operationId": "list_ingested_v1_ingest_list_get",
        "responses": {
          "200": {
            "description": "Successful Response",
            "content": {
              "application/json": {
                "schema": {
                  "$ref": "#/components/schemas/IngestResponse"
                }
              }
            }
          }
        }
      }
    },
    "/v1/ingest/{doc_id}": {
      "delete": {
        "tags": [
          "Ingestion"
        ],
        "summary": "Delete Ingested",
        "description": "Delete the specified ingested Document.\n\nThe `doc_id` can be obtained from the `GET /ingest/list` endpoint.\nThe document will be effectively deleted from your storage context.",
        "operationId": "delete_ingested_v1_ingest__doc_id__delete",
        "parameters": [
          {
            "name": "doc_id",
            "in": "path",
            "required": true,
            "schema": {
              "type": "string",
              "title": "Doc Id"
            }
          }
        ],
        "responses": {
          "200": {
            "description": "Successful Response",
            "content": {
              "application/json": {
                "schema": {}
              }
            }
          },
          "422": {
            "description": "Validation Error",
            "content": {
              "application/json": {
                "schema": {
                  "$ref": "#/components/schemas/HTTPValidationError"
                }
              }
            }
          }
        }
      }
    },
    "/v1/embeddings": {
      "post": {
        "tags": [
          "Embeddings"
        ],
        "summary": "Embeddings Generation",
        "description": "Get a vector representation of a given input.\n\nThat vector representation can be easily consumed\nby machine learning models and algorithms.",
        "operationId": "embeddings_generation_v1_embeddings_post",
        "requestBody": {
          "content": {
            "application/json": {
              "schema": {
                "$ref": "#/components/schemas/EmbeddingsBody"
              }
            }
          },
          "required": true
        },
        "responses": {
          "200": {
            "description": "Successful Response",
            "content": {
              "application/json": {
                "schema": {
                  "$ref": "#/components/schemas/EmbeddingsResponse"
                }
              }
            }
          },
          "422": {
            "description": "Validation Error",
            "content": {
              "application/json": {
                "schema": {
                  "$ref": "#/components/schemas/HTTPValidationError"
                }
              }
            }
          }
        }
      }
    },
    "/health": {
      "get": {
        "tags": [
          "Health"
        ],
        "summary": "Health",
        "description": "Return ok if the system is up.",
        "operationId": "health_health_get",
        "responses": {
          "200": {
            "description": "Successful Response",
            "content": {
              "application/json": {
                "schema": {
                  "$ref": "#/components/schemas/HealthResponse"
                }
              }
            }
          }
        }
      }
    }
  },
  "components": {
    "schemas": {
      "Body_ingest_v1_ingest_post": {
        "properties": {
          "file": {
            "type": "string",
            "format": "binary",
            "title": "File"
          }
        },
        "type": "object",
        "required": [
          "file"
        ],
        "title": "Body_ingest_v1_ingest_post"
      },
      "ChatBody": {
        "properties": {
          "messages": {
            "items": {
              "$ref": "#/components/schemas/OpenAIMessage"
            },
            "type": "array",
            "title": "Messages"
          },
          "use_context": {
            "type": "boolean",
            "title": "Use Context",
            "default": false
          },
          "context_filter": {
            "anyOf": [
              {
                "$ref": "#/components/schemas/ContextFilter"
              },
              {
                "type": "null"
              }
            ]
          },
          "include_sources": {
            "type": "boolean",
            "title": "Include Sources",
            "default": true
          },
          "stream": {
            "type": "boolean",
            "title": "Stream",
            "default": false
          }
        },
        "type": "object",
        "required": [
          "messages"
        ],
        "title": "ChatBody",
        "examples": [
          {
            "context_filter": {
              "docs_ids": [
                "c202d5e6-7b69-4869-81cc-dd574ee8ee11"
              ]
            },
            "include_sources": true,
            "messages": [
              {
                "content": "How do you fry an egg?",
                "role": "user"
              }
            ],
            "stream": false,
            "use_context": true
          }
        ]
      },
      "Chunk": {
        "properties": {
          "object": {
            "const": "context.chunk",
            "title": "Object"
          },
          "score": {
            "type": "number",
            "title": "Score",
            "examples": [
              0.023
            ]
          },
          "document": {
            "$ref": "#/components/schemas/IngestedDoc"
          },
          "text": {
            "type": "string",
            "title": "Text",
            "examples": [
              "Outbound sales increased 20%, driven by new leads."
            ]
          },
          "previous_texts": {
            "anyOf": [
              {
                "items": {
                  "type": "string"
                },
                "type": "array"
              },
              {
                "type": "null"
              }
            ],
            "title": "Previous Texts",
            "examples": [
              [
                "SALES REPORT 2023",
                "Inbound didn't show major changes."
              ]
            ]
          },
          "next_texts": {
            "anyOf": [
              {
                "items": {
                  "type": "string"
                },
                "type": "array"
              },
              {
                "type": "null"
              }
            ],
            "title": "Next Texts",
            "examples": [
              [
                "New leads came from Google Ads campaign.",
                "The campaign was run by the Marketing Department"
              ]
            ]
          }
        },
        "type": "object",
        "required": [
          "object",
          "score",
          "document",
          "text"
        ],
        "title": "Chunk"
      },
      "ChunksBody": {
        "properties": {
          "text": {
            "type": "string",
            "title": "Text",
            "examples": [
              "Q3 2023 sales"
            ]
          },
          "context_filter": {
            "anyOf": [
              {
                "$ref": "#/components/schemas/ContextFilter"
              },
              {
                "type": "null"
              }
            ]
          },
          "limit": {
            "type": "integer",
            "title": "Limit",
            "default": 10
          },
          "prev_next_chunks": {
            "type": "integer",
            "title": "Prev Next Chunks",
            "default": 0,
            "examples": [
              2
            ]
          }
        },
        "type": "object",
        "required": [
          "text"
        ],
        "title": "ChunksBody"
      },
      "ChunksResponse": {
        "properties": {
          "object": {
            "const": "list",
            "title": "Object"
          },
          "model": {
            "const": "private-gpt",
            "title": "Model"
          },
          "data": {
            "items": {
              "$ref": "#/components/schemas/Chunk"
            },
            "type": "array",
            "title": "Data"
          }
        },
        "type": "object",
        "required": [
          "object",
          "model",
          "data"
        ],
        "title": "ChunksResponse"
      },
      "CompletionsBody": {
        "properties": {
          "prompt": {
            "type": "string",
            "title": "Prompt"
          },
          "use_context": {
            "type": "boolean",
            "title": "Use Context",
            "default": false
          },
          "context_filter": {
            "anyOf": [
              {
                "$ref": "#/components/schemas/ContextFilter"
              },
              {
                "type": "null"
              }
            ]
          },
          "include_sources": {
            "type": "boolean",
            "title": "Include Sources",
            "default": true
          },
          "stream": {
            "type": "boolean",
            "title": "Stream",
            "default": false
          }
        },
        "type": "object",
        "required": [
          "prompt"
        ],
        "title": "CompletionsBody",
        "examples": [
          {
            "include_sources": false,
            "prompt": "How do you fry an egg?",
            "stream": false,
            "use_context": false
          }
        ]
      },
      "ContextFilter": {
        "properties": {
          "docs_ids": {
            "anyOf": [
              {
                "items": {
                  "type": "string"
                },
                "type": "array"
              },
              {
                "type": "null"
              }
            ],
            "title": "Docs Ids",
            "examples": [
              [
                "c202d5e6-7b69-4869-81cc-dd574ee8ee11"
              ]
            ]
          }
        },
        "type": "object",
        "required": [
          "docs_ids"
        ],
        "title": "ContextFilter"
      },
      "Embedding": {
        "properties": {
          "index": {
            "type": "integer",
            "title": "Index"
          },
          "object": {
            "const": "embedding",
            "title": "Object"
          },
          "embedding": {
            "items": {
              "type": "number"
            },
            "type": "array",
            "title": "Embedding",
            "examples": [
              [
                0.0023064255,
                -0.009327292
              ]
            ]
          }
        },
        "type": "object",
        "required": [
          "index",
          "object",
          "embedding"
        ],
        "title": "Embedding"
      },
      "EmbeddingsBody": {
        "properties": {
          "input": {
            "anyOf": [
              {
                "type": "string"
              },
              {
                "items": {
                  "type": "string"
                },
                "type": "array"
              }
            ],
            "title": "Input"
          }
        },
        "type": "object",
        "required": [
          "input"
        ],
        "title": "EmbeddingsBody"
      },
      "EmbeddingsResponse": {
        "properties": {
          "object": {
            "const": "list",
            "title": "Object"
          },
          "model": {
            "const": "private-gpt",
            "title": "Model"
          },
          "data": {
            "items": {
              "$ref": "#/components/schemas/Embedding"
            },
            "type": "array",
            "title": "Data"
          }
        },
        "type": "object",
        "required": [
          "object",
          "model",
          "data"
        ],
        "title": "EmbeddingsResponse"
      },
      "HTTPValidationError": {
        "properties": {
          "detail": {
            "items": {
              "$ref": "#/components/schemas/ValidationError"
            },
            "type": "array",
            "title": "Detail"
          }
        },
        "type": "object",
        "title": "HTTPValidationError"
      },
      "HealthResponse": {
        "properties": {
          "status": {
            "const": "ok",
            "title": "Status",
            "default": "ok"
          }
        },
        "type": "object",
        "title": "HealthResponse"
      },
      "IngestResponse": {
        "properties": {
          "object": {
            "const": "list",
            "title": "Object"
          },
          "model": {
            "const": "private-gpt",
            "title": "Model"
          },
          "data": {
            "items": {
              "$ref": "#/components/schemas/IngestedDoc"
            },
            "type": "array",
            "title": "Data"
          }
        },
        "type": "object",
        "required": [
          "object",
          "model",
          "data"
        ],
        "title": "IngestResponse"
      },
      "IngestedDoc": {
        "properties": {
          "object": {
            "const": "ingest.document",
            "title": "Object"
          },
          "doc_id": {
            "type": "string",
            "title": "Doc Id",
            "examples": [
              "c202d5e6-7b69-4869-81cc-dd574ee8ee11"
            ]
          },
          "doc_metadata": {
            "anyOf": [
              {
                "type": "object"
              },
              {
                "type": "null"
              }
            ],
            "title": "Doc Metadata",
            "examples": [
              {
                "file_name": "Sales Report Q3 2023.pdf",
                "page_label": "2"
              }
            ]
          }
        },
        "type": "object",
        "required": [
          "object",
          "doc_id",
          "doc_metadata"
        ],
        "title": "IngestedDoc"
      },
      "OpenAIChoice": {
        "properties": {
          "finish_reason": {
            "anyOf": [
              {
                "type": "string"
              },
              {
                "type": "null"
              }
            ],
            "title": "Finish Reason",
            "examples": [
              "stop"
            ]
          },
          "delta": {
            "anyOf": [
              {
                "$ref": "#/components/schemas/OpenAIDelta"
              },
              {
                "type": "null"
              }
            ]
          },
          "message": {
            "anyOf": [
              {
                "$ref": "#/components/schemas/OpenAIMessage"
              },
              {
                "type": "null"
              }
            ]
          },
          "sources": {
            "anyOf": [
              {
                "items": {
                  "$ref": "#/components/schemas/Chunk"
                },
                "type": "array"
              },
              {
                "type": "null"
              }
            ],
            "title": "Sources"
          },
          "index": {
            "type": "integer",
            "title": "Index",
            "default": 0
          }
        },
        "type": "object",
        "required": [
          "finish_reason"
        ],
        "title": "OpenAIChoice",
        "description": "Response from AI.\n\nEither the delta or the message will be present, but never both.\nSources used will be returned in case context retrieval was enabled."
      },
      "OpenAICompletion": {
        "properties": {
          "id": {
            "type": "string",
            "title": "Id"
          },
          "object": {
            "type": "string",
            "enum": [
              "completion",
              "completion.chunk"
            ],
            "title": "Object",
            "default": "completion"
          },
          "created": {
            "type": "integer",
            "title": "Created",
            "examples": [
              1623340000
            ]
          },
          "model": {
            "const": "private-gpt",
            "title": "Model"
          },
          "choices": {
            "items": {
              "$ref": "#/components/schemas/OpenAIChoice"
            },
            "type": "array",
            "title": "Choices"
          }
        },
        "type": "object",
        "required": [
          "id",
          "created",
          "model",
          "choices"
        ],
        "title": "OpenAICompletion",
        "description": "Clone of OpenAI Completion model.\n\nFor more information see: https://platform.openai.com/docs/api-reference/chat/object"
      },
      "OpenAIDelta": {
        "properties": {
          "content": {
            "anyOf": [
              {
                "type": "string"
              },
              {
                "type": "null"
              }
            ],
            "title": "Content"
          }
        },
        "type": "object",
        "required": [
          "content"
        ],
        "title": "OpenAIDelta",
        "description": "A piece of completion that needs to be concatenated to get the full message."
      },
      "OpenAIMessage": {
        "properties": {
          "role": {
            "type": "string",
            "enum": [
              "assistant",
              "system",
              "user"
            ],
            "title": "Role",
            "default": "user"
          },
          "content": {
            "anyOf": [
              {
                "type": "string"
              },
              {
                "type": "null"
              }
            ],
            "title": "Content"
          }
        },
        "type": "object",
        "required": [
          "content"
        ],
        "title": "OpenAIMessage",
        "description": "Inference result, with the source of the message.\n\nRole could be the assistant or system\n(providing a default response, not AI generated)."
      },
      "ValidationError": {
        "properties": {
          "loc": {
            "items": {
              "anyOf": [
                {
                  "type": "string"
                },
                {
                  "type": "integer"
                }
              ]
            },
            "type": "array",
            "title": "Location"
          },
          "msg": {
            "type": "string",
            "title": "Message"
          },
          "type": {
            "type": "string",
            "title": "Error Type"
          }
        },
        "type": "object",
        "required": [
          "loc",
          "msg",
          "type"
        ],
        "title": "ValidationError"
      }
    }
  },
  "tags": [
    {
      "name": "Ingestion",
      "description": "High-level APIs covering document ingestion -internally managing document parsing, splitting,metadata extraction, embedding generation and storage- and ingested documents CRUD.Each ingested document is identified by an ID that can be used to filter the contextused in *Contextual Completions* and *Context Chunks* APIs."
    },
    {
      "name": "Contextual Completions",
      "description": "High-level APIs covering contextual Chat and Completions. They follow OpenAI's format, extending it to allow using the context coming from ingested documents to create the response. Internallymanage context retrieval, prompt engineering and the response generation."
    },
    {
      "name": "Context Chunks",
      "description": "Low-level API that given a query return relevant chunks of text coming from the ingesteddocuments."
    },
    {
      "name": "Embeddings",
      "description": "Low-level API to obtain the vector representation of a given text, using an Embeddings model.Follows OpenAI's embeddings API format."
    },
    {
      "name": "Health",
      "description": "Simple health API to make sure the server is up and running."
    }
  ]
}