devupai 1.0.0a1__tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (42) hide show
  1. devupai-1.0.0a1/.gitignore +25 -0
  2. devupai-1.0.0a1/CHANGELOG.md +42 -0
  3. devupai-1.0.0a1/LICENSE +21 -0
  4. devupai-1.0.0a1/PKG-INFO +221 -0
  5. devupai-1.0.0a1/README.md +193 -0
  6. devupai-1.0.0a1/pyproject.toml +78 -0
  7. devupai-1.0.0a1/src/devupai/__init__.py +85 -0
  8. devupai-1.0.0a1/src/devupai/_client.py +1154 -0
  9. devupai-1.0.0a1/src/devupai/_errors.py +314 -0
  10. devupai-1.0.0a1/src/devupai/_response.py +225 -0
  11. devupai-1.0.0a1/src/devupai/_streaming.py +515 -0
  12. devupai-1.0.0a1/src/devupai/_version.py +3 -0
  13. devupai-1.0.0a1/src/devupai/py.typed +0 -0
  14. devupai-1.0.0a1/src/devupai/resources/__init__.py +137 -0
  15. devupai-1.0.0a1/src/devupai/resources/_base.py +207 -0
  16. devupai-1.0.0a1/src/devupai/resources/audio.py +663 -0
  17. devupai-1.0.0a1/src/devupai/resources/balance.py +103 -0
  18. devupai-1.0.0a1/src/devupai/resources/chat.py +428 -0
  19. devupai-1.0.0a1/src/devupai/resources/embeddings.py +141 -0
  20. devupai-1.0.0a1/src/devupai/resources/health.py +113 -0
  21. devupai-1.0.0a1/src/devupai/resources/images.py +503 -0
  22. devupai-1.0.0a1/src/devupai/resources/inference.py +347 -0
  23. devupai-1.0.0a1/src/devupai/resources/me.py +208 -0
  24. devupai-1.0.0a1/src/devupai/resources/models.py +405 -0
  25. devupai-1.0.0a1/src/devupai/resources/rerank.py +147 -0
  26. devupai-1.0.0a1/src/devupai/resources/responses.py +400 -0
  27. devupai-1.0.0a1/src/devupai/resources/shared.py +314 -0
  28. devupai-1.0.0a1/src/devupai/resources/video.py +600 -0
  29. devupai-1.0.0a1/src/devupai/types/__init__.py +264 -0
  30. devupai-1.0.0a1/src/devupai/types/_base.py +167 -0
  31. devupai-1.0.0a1/src/devupai/types/account.py +48 -0
  32. devupai-1.0.0a1/src/devupai/types/audio.py +89 -0
  33. devupai-1.0.0a1/src/devupai/types/chat.py +284 -0
  34. devupai-1.0.0a1/src/devupai/types/embeddings.py +51 -0
  35. devupai-1.0.0a1/src/devupai/types/images.py +238 -0
  36. devupai-1.0.0a1/src/devupai/types/models.py +107 -0
  37. devupai-1.0.0a1/src/devupai/types/rerank.py +41 -0
  38. devupai-1.0.0a1/src/devupai/types/responses.py +219 -0
  39. devupai-1.0.0a1/src/devupai/types/shared.py +79 -0
  40. devupai-1.0.0a1/src/devupai/types/video.py +218 -0
  41. devupai-1.0.0a1/src/devupai/types/webhooks.py +61 -0
  42. devupai-1.0.0a1/src/devupai/webhooks.py +278 -0
@@ -0,0 +1,25 @@
1
+ # Virtual environments
2
+ .venv/
3
+ .venv314/
4
+ .venv*/
5
+
6
+ # Bytecode and tool state folders
7
+ __pycache__/
8
+ *.py[cod]
9
+ .pytest_*/
10
+ .mypy_*/
11
+ .ruff_*/
12
+
13
+ # Coverage
14
+ .coverage
15
+ .coverage.*
16
+ htmlcov/
17
+
18
+ # Build output
19
+ dist/
20
+ build/
21
+ *.egg-info/
22
+
23
+ # Local settings
24
+ .env
25
+ .env.*
@@ -0,0 +1,42 @@
1
+ # Changelog
2
+
3
+ All notable changes to the `devupai` Python package are documented in this file.
4
+
5
+ ## [1.0.0a1]
6
+
7
+ The first pre-release of the official Python SDK for the DEVUP AI API, with a
8
+ synchronous client (`DevupAI`) and an asynchronous client (`AsyncDevupAI`).
9
+ It needs Python 3.11 or later. Install it with `pip install --pre devupai`.
10
+
11
+ ### Added
12
+
13
+ - `client.chat.completions.create()` and `client.responses.create()`, with
14
+ JSON responses and `stream=True`. A stream is an iterator (`async for` on
15
+ `AsyncDevupAI`) that also works in a `with` block, and its `request_id` and
16
+ `warnings` are readable before the first chunk.
17
+ - `client.models.list()` and `client.models.retrieve()`.
18
+ - `client.images.generate()`, `client.images.edit()` and
19
+ `client.images.download()`.
20
+ - `client.embeddings.create()` and `client.rerank.create()`.
21
+ - `client.audio.speech.create()` and `client.audio.transcriptions.create()`.
22
+ - `client.video.generations.create()` and `client.video.edits.create()`.
23
+ - `client.inference.run()`, with JSON or binary results.
24
+ - `client.balance.retrieve()`, `client.me.retrieve()`, `client.me.avatar()`
25
+ and `client.health.check()`.
26
+ - File uploads from `bytes`, a binary file object, a `pathlib.Path` or a
27
+ `(filename, content, content_type)` tuple. The SDK closes the files it opens
28
+ from a path and never closes a file object you pass.
29
+ - `BinaryResponse` and `AsyncBinaryResponse` for binary results, with
30
+ `read()`, `iter_bytes()` and `write_to_file()`.
31
+ - `with_response` on each resource that calls the API, to read the status
32
+ code, headers and `request_id` together with the parsed result.
33
+ - Webhook signature verification: `client.webhooks.verify_signature()`,
34
+ `client.webhooks.construct_event()`, `verify_webhook_signature()` and
35
+ `construct_webhook_event()`.
36
+ - Error classes for each HTTP status the API documents, all subclasses of
37
+ `DevupAPIError`, plus `APIConnectionError` and `APITimeoutError`.
38
+ `error.code` names the reason, and `error.request_id` identifies the request.
39
+ - Retries with backoff for requests that are safe to send again. A billed
40
+ request is sent again only after a 429 or 503 answer, or when the
41
+ connection could not be opened.
42
+ - Type hints for every request and response (`py.typed`).
@@ -0,0 +1,21 @@
1
+ MIT License
2
+
3
+ Copyright (c) 2026 DEVUP AI
4
+
5
+ Permission is hereby granted, free of charge, to any person obtaining a copy
6
+ of this software and associated documentation files (the "Software"), to deal
7
+ in the Software without restriction, including without limitation the rights
8
+ to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
9
+ copies of the Software, and to permit persons to whom the Software is
10
+ furnished to do so, subject to the following conditions:
11
+
12
+ The above copyright notice and this permission notice shall be included in all
13
+ copies or substantial portions of the Software.
14
+
15
+ THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
16
+ IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
17
+ FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
18
+ AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
19
+ LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
20
+ OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
21
+ SOFTWARE.
@@ -0,0 +1,221 @@
1
+ Metadata-Version: 2.5
2
+ Name: devupai
3
+ Version: 1.0.0a1
4
+ Summary: Official Python SDK for the DEVUP AI API, with billing in DZD.
5
+ Project-URL: Homepage, https://github.com/devupai-platform/devupai-python#readme
6
+ Project-URL: Repository, https://github.com/devupai-platform/devupai-python
7
+ Project-URL: Issues, https://github.com/devupai-platform/devupai-python/issues
8
+ Project-URL: Changelog, https://github.com/devupai-platform/devupai-python/blob/main/CHANGELOG.md
9
+ Author: DEVUP AI
10
+ License-Expression: MIT
11
+ License-File: LICENSE
12
+ Keywords: ai,api,devup,devupai,dzd,sdk
13
+ Classifier: Development Status :: 3 - Alpha
14
+ Classifier: Intended Audience :: Developers
15
+ Classifier: Operating System :: OS Independent
16
+ Classifier: Programming Language :: Python :: 3
17
+ Classifier: Programming Language :: Python :: 3 :: Only
18
+ Classifier: Programming Language :: Python :: 3.11
19
+ Classifier: Programming Language :: Python :: 3.12
20
+ Classifier: Programming Language :: Python :: 3.13
21
+ Classifier: Programming Language :: Python :: 3.14
22
+ Classifier: Programming Language :: Python :: 3.15
23
+ Classifier: Typing :: Typed
24
+ Requires-Python: >=3.11
25
+ Requires-Dist: httpx<1,>=0.28.1
26
+ Requires-Dist: pydantic<3,>=2.10
27
+ Description-Content-Type: text/markdown
28
+
29
+ # DEVUP AI Python SDK
30
+
31
+ The official Python SDK for the DEVUP AI API, with a synchronous client,
32
+ `DevupAI`, and an asynchronous client, `AsyncDevupAI`.
33
+
34
+ ## Requirements
35
+
36
+ Python 3.11 or later.
37
+
38
+ ## Installation
39
+
40
+ ```bash
41
+ pip install --pre devupai
42
+ ```
43
+
44
+ Version 1.0.0 removes the need for `--pre`.
45
+
46
+ ## Create a client
47
+
48
+ Synchronous client:
49
+
50
+ ```python
51
+ from devupai import DevupAI
52
+
53
+ with DevupAI() as client:
54
+ print(client.base_url)
55
+ ```
56
+
57
+ Asynchronous client:
58
+
59
+ ```python
60
+ import asyncio
61
+
62
+ from devupai import AsyncDevupAI
63
+
64
+
65
+ async def main() -> None:
66
+ async with AsyncDevupAI() as client:
67
+ print(client.base_url)
68
+
69
+
70
+ asyncio.run(main())
71
+ ```
72
+
73
+ You can also pass the key yourself:
74
+
75
+ ```python
76
+ from devupai import DevupAI
77
+
78
+ client = DevupAI(api_key="your-api-key")
79
+ try:
80
+ print(client.base_url)
81
+ finally:
82
+ client.close()
83
+ ```
84
+
85
+ Without a `with` block, call `client.close()` (or `await client.close()` for
86
+ `AsyncDevupAI`) when you are done. Closing a client more than once is allowed.
87
+
88
+ ## API key
89
+
90
+ The client takes the API key from the first of these that is set:
91
+
92
+ 1. The `api_key` argument.
93
+ 2. The `DEVUP_API_KEY` environment variable.
94
+ 3. The `DEVUPAI_API_KEY` environment variable. It is read only when
95
+ `DEVUP_API_KEY` is not set at all.
96
+
97
+ When none of them is set, or the environment variable is empty, creating the
98
+ client raises `ValueError`. A key that contains a line break or a NUL
99
+ character also raises `ValueError`.
100
+
101
+ `api_key=""` is allowed: the client then sends an empty key, which is useful
102
+ only for requests that do not need one.
103
+
104
+ ## Options
105
+
106
+ Every option is a keyword argument, and both clients take the same options.
107
+
108
+ ```python
109
+ from devupai import DevupAI
110
+
111
+ client = DevupAI(
112
+ api_key="your-api-key",
113
+ base_url="https://api.devupai.com/v1",
114
+ timeout=60.0,
115
+ max_retries=2,
116
+ default_headers={"X-Application-Name": "my-application"},
117
+ )
118
+ ```
119
+
120
+ | Option | Type | Default | Description |
121
+ | ----------------- | --------------------------------------- | ---------------------------- | ----------- |
122
+ | `api_key` | `str \| None` | See [API key](#api-key) | Your DEVUP AI API key. |
123
+ | `base_url` | `str \| None` | `https://api.devupai.com/v1` | DEVUP AI API base URL. A trailing `/` is removed. It must be an absolute `http` or `https` URL without a user name or password (`ValueError` otherwise). |
124
+ | `timeout` | `float \| None` | `600.0` (10 minutes) | Seconds. One limit for the whole call, retries included. The Node.js SDK takes this value in milliseconds; this SDK takes seconds. |
125
+ | `max_retries` | `int \| None` | `2` | How many times a request is sent again. See [Retries](#retries). |
126
+ | `default_headers` | `Mapping[str, str] \| None` | — | Headers included in every request. |
127
+ | `http_client` | `httpx.Client` / `httpx.AsyncClient` | A client the SDK creates | Your own `httpx.Client` (for `DevupAI`) or `httpx.AsyncClient` (for `AsyncDevupAI`). Only its transport, connection pool and event hooks apply: its own `auth`, base URL, headers, query parameters, cookies and timeouts are not used (each request takes the `timeout` of this client or of the call). The SDK never closes a client you pass. |
128
+ | `on_debug` | `Callable[[DebugEvent], None] \| None` | — | Receives debug events, such as `retry` before each retry. An exception raised in it is ignored. |
129
+
130
+ `timeout` must be a positive number of seconds, and `max_retries` a
131
+ non-negative integer. A wrong type raises `TypeError` and a wrong value raises
132
+ `ValueError`, before any request is sent.
133
+
134
+ The SDK sends `User-Agent: devupai-python/<version>`. It manages the
135
+ `Authorization` and `Content-Type` headers itself; custom headers cannot
136
+ override them. Redirects are not followed.
137
+
138
+ ## Retries
139
+
140
+ The client sends a request again, up to `max_retries` times (default 2), after
141
+ a 408, 429, 502, 503 or 504 answer, or a connection error.
142
+
143
+ | Requests | Sent again after |
144
+ | -------- | ---------------- |
145
+ | Billed requests: chat completions, Responses, embeddings, rerank, image generation and edits, audio, video, and native inference | 429 and 503, and a connection that could not be opened |
146
+ | Other requests | 408, 429, 502, 503, and 504, and connection errors |
147
+
148
+ A billed request is sent again only after a 429 or 503 answer, or when the
149
+ connection could not be opened. After any other answer, or a connection lost
150
+ after the request may have been sent, the error is returned to you without a
151
+ retry, because the request may already have been processed and charged.
152
+
153
+ A 500 answer is not retried.
154
+
155
+ When the answer has a `retry-after-ms` or `Retry-After` header, the client
156
+ waits that long before the next attempt. Otherwise it waits a random time that
157
+ grows with each attempt, at most 60 seconds between attempts. A wait that does
158
+ not fit in the time left of `timeout` ends the call with the error.
159
+
160
+ `max_retries=0` turns retries off.
161
+
162
+ ## Errors
163
+
164
+ Every error from the API is a `DevupAPIError`. The SDK raises the subclass
165
+ that matches the HTTP status, so you can catch a specific case or the base
166
+ class. All the classes below are subclasses of `DevupAPIError`.
167
+
168
+ | Class | Status | Notes |
169
+ | --------------------------- | ---------- | ----- |
170
+ | `BadRequestError` | 400 | The request is not valid as sent; `code` names the reason. |
171
+ | `AuthenticationError` | 401 | Missing or invalid API key. |
172
+ | `PaymentRequiredError` | 402 | The balance does not cover the request (`insufficient_balance`), or a scoped credential reached its budget (`budget_exhausted`). |
173
+ | `PermissionDeniedError` | 403 | The key cannot use this resource. |
174
+ | `NotFoundError` | 404 | For example, an unknown model. |
175
+ | `ConflictError` | 409 | The outcome of this request is not confirmed yet (`request_outcome_pending`). Do not send it again: contact support with `request_id`. |
176
+ | `PayloadTooLargeError` | 413 | The request body is larger than the limit (`payload_too_large`). |
177
+ | `UnsupportedMediaTypeError` | 415 | The model does not accept the content type of the request. Send the `Content-Type` and body format that the API reference lists for the endpoint. |
178
+ | `UnprocessableEntityError` | 422 | The request was not accepted in its current form. |
179
+ | `RateLimitError` | 429 | Read `retry_after_ms` for the suggested wait. |
180
+ | `InternalServerError` | 500 and up | After a 500 or 502 the request may already have been processed and charged: check its outcome before sending it again. |
181
+ | `APIConnectionError` | `0` | The API could not be reached. |
182
+ | `APITimeoutError` | `0` | The call exceeded `timeout`. A subclass of `APIConnectionError`. |
183
+ | `DevupAPIError` | other | Any other status. |
184
+
185
+ | Field | Type | Description |
186
+ | ------------------- | ---------------- | ----------- |
187
+ | `status` | `int` | HTTP status code, or `0` for connection errors and timeouts. |
188
+ | `message` | `str` | Human-readable error message (also `str(err)`). |
189
+ | `code` | `str \| None` | Machine-readable error code, when available. |
190
+ | `type` | `str \| None` | Error category, when available. |
191
+ | `param` | `str \| None` | The request parameter the error refers to, when available. |
192
+ | `request_id` | `str \| None` | The `x-request-id` header. Include it when you contact support. |
193
+ | `client_request_id` | `str \| None` | The `x-client-request-id` header. |
194
+ | `retry_after` | `str \| None` | Raw `Retry-After` header value. |
195
+ | `retry_after_ms` | `float \| None` | Suggested wait in milliseconds, when the answer gives one. |
196
+ | `warnings` | `list[str]` | Notice codes from the `X-DEVUP-Warnings` header (empty when none). |
197
+ | `headers` | `dict[str, str]` | Response headers, with lower-case names. |
198
+
199
+ The underlying exception, when there is one, is chained as `__cause__`.
200
+
201
+ A 409 means the outcome of the request is not confirmed yet. Do not send it
202
+ again: contact support with its `request_id`.
203
+
204
+ ```python
205
+ from devupai import DevupAPIError, RateLimitError
206
+
207
+ try:
208
+ ... # a call to the DEVUP AI API
209
+ except RateLimitError as err:
210
+ print(err.retry_after_ms)
211
+ except DevupAPIError as err:
212
+ print(err.status, err.code, err.request_id)
213
+ ```
214
+
215
+ The error codes you can act on are listed at
216
+ [docs.devupai.com/docs/errors](https://docs.devupai.com/docs/errors).
217
+
218
+ ## License
219
+
220
+ The DEVUP AI Python SDK is released under the
221
+ [MIT License](https://github.com/devupai-platform/devupai-python/blob/main/LICENSE).
@@ -0,0 +1,193 @@
1
+ # DEVUP AI Python SDK
2
+
3
+ The official Python SDK for the DEVUP AI API, with a synchronous client,
4
+ `DevupAI`, and an asynchronous client, `AsyncDevupAI`.
5
+
6
+ ## Requirements
7
+
8
+ Python 3.11 or later.
9
+
10
+ ## Installation
11
+
12
+ ```bash
13
+ pip install --pre devupai
14
+ ```
15
+
16
+ Version 1.0.0 removes the need for `--pre`.
17
+
18
+ ## Create a client
19
+
20
+ Synchronous client:
21
+
22
+ ```python
23
+ from devupai import DevupAI
24
+
25
+ with DevupAI() as client:
26
+ print(client.base_url)
27
+ ```
28
+
29
+ Asynchronous client:
30
+
31
+ ```python
32
+ import asyncio
33
+
34
+ from devupai import AsyncDevupAI
35
+
36
+
37
+ async def main() -> None:
38
+ async with AsyncDevupAI() as client:
39
+ print(client.base_url)
40
+
41
+
42
+ asyncio.run(main())
43
+ ```
44
+
45
+ You can also pass the key yourself:
46
+
47
+ ```python
48
+ from devupai import DevupAI
49
+
50
+ client = DevupAI(api_key="your-api-key")
51
+ try:
52
+ print(client.base_url)
53
+ finally:
54
+ client.close()
55
+ ```
56
+
57
+ Without a `with` block, call `client.close()` (or `await client.close()` for
58
+ `AsyncDevupAI`) when you are done. Closing a client more than once is allowed.
59
+
60
+ ## API key
61
+
62
+ The client takes the API key from the first of these that is set:
63
+
64
+ 1. The `api_key` argument.
65
+ 2. The `DEVUP_API_KEY` environment variable.
66
+ 3. The `DEVUPAI_API_KEY` environment variable. It is read only when
67
+ `DEVUP_API_KEY` is not set at all.
68
+
69
+ When none of them is set, or the environment variable is empty, creating the
70
+ client raises `ValueError`. A key that contains a line break or a NUL
71
+ character also raises `ValueError`.
72
+
73
+ `api_key=""` is allowed: the client then sends an empty key, which is useful
74
+ only for requests that do not need one.
75
+
76
+ ## Options
77
+
78
+ Every option is a keyword argument, and both clients take the same options.
79
+
80
+ ```python
81
+ from devupai import DevupAI
82
+
83
+ client = DevupAI(
84
+ api_key="your-api-key",
85
+ base_url="https://api.devupai.com/v1",
86
+ timeout=60.0,
87
+ max_retries=2,
88
+ default_headers={"X-Application-Name": "my-application"},
89
+ )
90
+ ```
91
+
92
+ | Option | Type | Default | Description |
93
+ | ----------------- | --------------------------------------- | ---------------------------- | ----------- |
94
+ | `api_key` | `str \| None` | See [API key](#api-key) | Your DEVUP AI API key. |
95
+ | `base_url` | `str \| None` | `https://api.devupai.com/v1` | DEVUP AI API base URL. A trailing `/` is removed. It must be an absolute `http` or `https` URL without a user name or password (`ValueError` otherwise). |
96
+ | `timeout` | `float \| None` | `600.0` (10 minutes) | Seconds. One limit for the whole call, retries included. The Node.js SDK takes this value in milliseconds; this SDK takes seconds. |
97
+ | `max_retries` | `int \| None` | `2` | How many times a request is sent again. See [Retries](#retries). |
98
+ | `default_headers` | `Mapping[str, str] \| None` | — | Headers included in every request. |
99
+ | `http_client` | `httpx.Client` / `httpx.AsyncClient` | A client the SDK creates | Your own `httpx.Client` (for `DevupAI`) or `httpx.AsyncClient` (for `AsyncDevupAI`). Only its transport, connection pool and event hooks apply: its own `auth`, base URL, headers, query parameters, cookies and timeouts are not used (each request takes the `timeout` of this client or of the call). The SDK never closes a client you pass. |
100
+ | `on_debug` | `Callable[[DebugEvent], None] \| None` | — | Receives debug events, such as `retry` before each retry. An exception raised in it is ignored. |
101
+
102
+ `timeout` must be a positive number of seconds, and `max_retries` a
103
+ non-negative integer. A wrong type raises `TypeError` and a wrong value raises
104
+ `ValueError`, before any request is sent.
105
+
106
+ The SDK sends `User-Agent: devupai-python/<version>`. It manages the
107
+ `Authorization` and `Content-Type` headers itself; custom headers cannot
108
+ override them. Redirects are not followed.
109
+
110
+ ## Retries
111
+
112
+ The client sends a request again, up to `max_retries` times (default 2), after
113
+ a 408, 429, 502, 503 or 504 answer, or a connection error.
114
+
115
+ | Requests | Sent again after |
116
+ | -------- | ---------------- |
117
+ | Billed requests: chat completions, Responses, embeddings, rerank, image generation and edits, audio, video, and native inference | 429 and 503, and a connection that could not be opened |
118
+ | Other requests | 408, 429, 502, 503, and 504, and connection errors |
119
+
120
+ A billed request is sent again only after a 429 or 503 answer, or when the
121
+ connection could not be opened. After any other answer, or a connection lost
122
+ after the request may have been sent, the error is returned to you without a
123
+ retry, because the request may already have been processed and charged.
124
+
125
+ A 500 answer is not retried.
126
+
127
+ When the answer has a `retry-after-ms` or `Retry-After` header, the client
128
+ waits that long before the next attempt. Otherwise it waits a random time that
129
+ grows with each attempt, at most 60 seconds between attempts. A wait that does
130
+ not fit in the time left of `timeout` ends the call with the error.
131
+
132
+ `max_retries=0` turns retries off.
133
+
134
+ ## Errors
135
+
136
+ Every error from the API is a `DevupAPIError`. The SDK raises the subclass
137
+ that matches the HTTP status, so you can catch a specific case or the base
138
+ class. All the classes below are subclasses of `DevupAPIError`.
139
+
140
+ | Class | Status | Notes |
141
+ | --------------------------- | ---------- | ----- |
142
+ | `BadRequestError` | 400 | The request is not valid as sent; `code` names the reason. |
143
+ | `AuthenticationError` | 401 | Missing or invalid API key. |
144
+ | `PaymentRequiredError` | 402 | The balance does not cover the request (`insufficient_balance`), or a scoped credential reached its budget (`budget_exhausted`). |
145
+ | `PermissionDeniedError` | 403 | The key cannot use this resource. |
146
+ | `NotFoundError` | 404 | For example, an unknown model. |
147
+ | `ConflictError` | 409 | The outcome of this request is not confirmed yet (`request_outcome_pending`). Do not send it again: contact support with `request_id`. |
148
+ | `PayloadTooLargeError` | 413 | The request body is larger than the limit (`payload_too_large`). |
149
+ | `UnsupportedMediaTypeError` | 415 | The model does not accept the content type of the request. Send the `Content-Type` and body format that the API reference lists for the endpoint. |
150
+ | `UnprocessableEntityError` | 422 | The request was not accepted in its current form. |
151
+ | `RateLimitError` | 429 | Read `retry_after_ms` for the suggested wait. |
152
+ | `InternalServerError` | 500 and up | After a 500 or 502 the request may already have been processed and charged: check its outcome before sending it again. |
153
+ | `APIConnectionError` | `0` | The API could not be reached. |
154
+ | `APITimeoutError` | `0` | The call exceeded `timeout`. A subclass of `APIConnectionError`. |
155
+ | `DevupAPIError` | other | Any other status. |
156
+
157
+ | Field | Type | Description |
158
+ | ------------------- | ---------------- | ----------- |
159
+ | `status` | `int` | HTTP status code, or `0` for connection errors and timeouts. |
160
+ | `message` | `str` | Human-readable error message (also `str(err)`). |
161
+ | `code` | `str \| None` | Machine-readable error code, when available. |
162
+ | `type` | `str \| None` | Error category, when available. |
163
+ | `param` | `str \| None` | The request parameter the error refers to, when available. |
164
+ | `request_id` | `str \| None` | The `x-request-id` header. Include it when you contact support. |
165
+ | `client_request_id` | `str \| None` | The `x-client-request-id` header. |
166
+ | `retry_after` | `str \| None` | Raw `Retry-After` header value. |
167
+ | `retry_after_ms` | `float \| None` | Suggested wait in milliseconds, when the answer gives one. |
168
+ | `warnings` | `list[str]` | Notice codes from the `X-DEVUP-Warnings` header (empty when none). |
169
+ | `headers` | `dict[str, str]` | Response headers, with lower-case names. |
170
+
171
+ The underlying exception, when there is one, is chained as `__cause__`.
172
+
173
+ A 409 means the outcome of the request is not confirmed yet. Do not send it
174
+ again: contact support with its `request_id`.
175
+
176
+ ```python
177
+ from devupai import DevupAPIError, RateLimitError
178
+
179
+ try:
180
+ ... # a call to the DEVUP AI API
181
+ except RateLimitError as err:
182
+ print(err.retry_after_ms)
183
+ except DevupAPIError as err:
184
+ print(err.status, err.code, err.request_id)
185
+ ```
186
+
187
+ The error codes you can act on are listed at
188
+ [docs.devupai.com/docs/errors](https://docs.devupai.com/docs/errors).
189
+
190
+ ## License
191
+
192
+ The DEVUP AI Python SDK is released under the
193
+ [MIT License](https://github.com/devupai-platform/devupai-python/blob/main/LICENSE).
@@ -0,0 +1,78 @@
1
+ [build-system]
2
+ requires = ["hatchling==1.33.0"]
3
+ build-backend = "hatchling.build"
4
+
5
+ [project]
6
+ name = "devupai"
7
+ dynamic = ["version"]
8
+ description = "Official Python SDK for the DEVUP AI API, with billing in DZD."
9
+ readme = "README.md"
10
+ license = "MIT"
11
+ license-files = ["LICENSE"]
12
+ requires-python = ">=3.11"
13
+ authors = [{ name = "DEVUP AI" }]
14
+ keywords = ["devupai", "devup", "ai", "api", "sdk", "dzd"]
15
+ dependencies = [
16
+ "httpx>=0.28.1,<1",
17
+ "pydantic>=2.10,<3",
18
+ ]
19
+ classifiers = [
20
+ "Development Status :: 3 - Alpha",
21
+ "Intended Audience :: Developers",
22
+ "Operating System :: OS Independent",
23
+ "Programming Language :: Python :: 3",
24
+ "Programming Language :: Python :: 3 :: Only",
25
+ "Programming Language :: Python :: 3.11",
26
+ "Programming Language :: Python :: 3.12",
27
+ "Programming Language :: Python :: 3.13",
28
+ "Programming Language :: Python :: 3.14",
29
+ "Programming Language :: Python :: 3.15",
30
+ "Typing :: Typed",
31
+ ]
32
+
33
+ [project.urls]
34
+ Homepage = "https://github.com/devupai-platform/devupai-python#readme"
35
+ Repository = "https://github.com/devupai-platform/devupai-python"
36
+ Issues = "https://github.com/devupai-platform/devupai-python/issues"
37
+ Changelog = "https://github.com/devupai-platform/devupai-python/blob/main/CHANGELOG.md"
38
+
39
+ [tool.hatch.version]
40
+ path = "src/devupai/_version.py"
41
+
42
+ [tool.hatch.build.targets.wheel]
43
+ packages = ["src/devupai"]
44
+
45
+ [tool.hatch.build.targets.sdist]
46
+ include = [
47
+ "src/devupai",
48
+ "README.md",
49
+ "CHANGELOG.md",
50
+ "LICENSE",
51
+ "pyproject.toml",
52
+ ]
53
+
54
+ [tool.pytest.ini_options]
55
+ testpaths = ["tests"]
56
+ pythonpath = ["src", "tests", "scripts"]
57
+ addopts = "-ra --strict-markers --strict-config"
58
+ xfail_strict = true
59
+
60
+ [tool.mypy]
61
+ strict = true
62
+ python_version = "3.11"
63
+ files = ["src", "scripts", "tests/type_checks"]
64
+
65
+ [tool.pyright]
66
+ include = ["src", "scripts", "tests/type_checks"]
67
+ typeCheckingMode = "strict"
68
+ pythonVersion = "3.11"
69
+
70
+ [tool.ruff]
71
+ target-version = "py311"
72
+ line-length = 120
73
+
74
+ [tool.ruff.lint]
75
+ select = ["E", "F", "W", "I", "UP", "B"]
76
+
77
+ [tool.ruff.lint.isort]
78
+ known-first-party = ["devupai"]
@@ -0,0 +1,85 @@
1
+ """The official Python client for the DEVUP AI API.
2
+
3
+ Create a client and call the API::
4
+
5
+ from devupai import DevupAI
6
+
7
+ client = DevupAI() # reads the key from DEVUP_API_KEY, then DEVUPAI_API_KEY
8
+
9
+ ``AsyncDevupAI`` offers the same options for ``asyncio`` code. ``create(stream=True)`` returns a ``Stream`` (an
10
+ ``AsyncStream`` on the async client), and a binary result is a ``BinaryResponse`` (``AsyncBinaryResponse``).
11
+
12
+ An answer or a connection problem raises a subclass of ``DevupAPIError`` (``APIConnectionError`` when no response
13
+ was received); a webhook that cannot be verified raises ``DevupWebhookVerificationError``. An option or header value
14
+ that is not valid raises ``TypeError`` or ``ValueError`` before anything is sent, and a closed client raises
15
+ ``RuntimeError``.
16
+
17
+ Request parameter and response types live in ``devupai.types``. Webhook signatures are checked with
18
+ ``client.webhooks``, or with ``verify_webhook_signature`` and ``construct_webhook_event`` without a client.
19
+ """
20
+
21
+ from devupai import types
22
+ from devupai._client import AsyncDevupAI, DebugEvent, DevupAI, RequestOptions
23
+ from devupai._errors import (
24
+ APIConnectionError,
25
+ APITimeoutError,
26
+ AuthenticationError,
27
+ BadRequestError,
28
+ ConflictError,
29
+ DevupAPIError,
30
+ DevupWebhookVerificationError,
31
+ InternalServerError,
32
+ NotFoundError,
33
+ PayloadTooLargeError,
34
+ PaymentRequiredError,
35
+ PermissionDeniedError,
36
+ RateLimitError,
37
+ UnprocessableEntityError,
38
+ UnsupportedMediaTypeError,
39
+ )
40
+ from devupai._response import APIResponse, AsyncBinaryResponse, BinaryResponse
41
+ from devupai._streaming import AsyncStream, Stream
42
+ from devupai._version import __version__
43
+ from devupai.webhooks import Webhooks, construct_webhook_event, verify_webhook_signature
44
+
45
+ VERSION: str = __version__
46
+ """The version of this package, the same value as ``__version__``."""
47
+
48
+ __all__ = [
49
+ # Clients and responses.
50
+ "DevupAI",
51
+ "AsyncDevupAI",
52
+ "APIResponse",
53
+ "DebugEvent",
54
+ "RequestOptions",
55
+ # Streams and binary results.
56
+ "Stream",
57
+ "AsyncStream",
58
+ "BinaryResponse",
59
+ "AsyncBinaryResponse",
60
+ # Version.
61
+ "VERSION",
62
+ "__version__",
63
+ # Errors.
64
+ "DevupAPIError",
65
+ "BadRequestError",
66
+ "AuthenticationError",
67
+ "PaymentRequiredError",
68
+ "PermissionDeniedError",
69
+ "NotFoundError",
70
+ "ConflictError",
71
+ "PayloadTooLargeError",
72
+ "UnsupportedMediaTypeError",
73
+ "UnprocessableEntityError",
74
+ "RateLimitError",
75
+ "InternalServerError",
76
+ "APIConnectionError",
77
+ "APITimeoutError",
78
+ "DevupWebhookVerificationError",
79
+ # Webhooks.
80
+ "Webhooks",
81
+ "verify_webhook_signature",
82
+ "construct_webhook_event",
83
+ # Request parameter and response types.
84
+ "types",
85
+ ]