aiperf 0.11.0.dev20260717__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.
@@ -0,0 +1,203 @@
1
+ <!--
2
+ SPDX-FileCopyrightText: Copyright (c) 2024-2025 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
3
+ SPDX-License-Identifier: Apache-2.0
4
+ -->
5
+
6
+ # Third-Party Asset Attributions
7
+
8
+ This document provides attribution information for public domain assets used in this project.
9
+
10
+ ## Text Assets
11
+
12
+ ### Shakespeare Complete Works
13
+
14
+ **Asset Information:**
15
+ - **File(s)**: [`aiperf/dataset/generator/assets/shakespeare.txt`](aiperf/dataset/generator/assets/shakespeare.txt)
16
+ - **Author**: William Shakespeare (1564-1616)
17
+ - **License**: Public Domain
18
+ - **Status**: Works created before copyright law; no restrictions apply
19
+
20
+ **License Text:**
21
+
22
+ > These works by William Shakespeare are in the public domain worldwide. No permission is required for any use.
23
+
24
+ ## Image Assets
25
+
26
+ ### Getty Museum Images
27
+
28
+ **Asset Information:**
29
+ - **Files**:
30
+ - [`aiperf/dataset/generator/assets/source_images/0bfd8fdf-457f-43c8-9253-a2346d37d26a_1024.jpg`](aiperf/dataset/generator/assets/source_images/0bfd8fdf-457f-43c8-9253-a2346d37d26a_1024.jpg) - [Object 103QTD](https://www.getty.edu/art/collection/object/103QTD)
31
+ - [`aiperf/dataset/generator/assets/source_images/119544eb-9bbf-47d1-8d93-a51de6370295_861.jpg`](aiperf/dataset/generator/assets/source_images/119544eb-9bbf-47d1-8d93-a51de6370295_861.jpg) - [Object 103SNG](https://www.getty.edu/art/collection/object/103SNG)
32
+ - [`aiperf/dataset/generator/assets/source_images/1ba066e0-7291-4ef4-8a6e-e8724af7f046_865.jpg`](aiperf/dataset/generator/assets/source_images/1ba066e0-7291-4ef4-8a6e-e8724af7f046_865.jpg) - [Object 10962M](https://www.getty.edu/art/collection/object/10962M)
33
+ - [`aiperf/dataset/generator/assets/source_images/c946f826-4a9a-4ca2-b891-9f791d38092b_1024.jpg`](aiperf/dataset/generator/assets/source_images/c946f826-4a9a-4ca2-b891-9f791d38092b_1024.jpg) - [Object 103RQ6](https://www.getty.edu/art/collection/object/103RQ6)
34
+ - **License**: CC0 1.0 Universal (Public Domain Dedication)
35
+ - **Attribution** (recommended): Digital image courtesy of Getty's Open Content Program
36
+
37
+ **License Text:**
38
+
39
+ > Creative Commons Legal Code
40
+ >
41
+ > CC0 1.0 Universal
42
+ >
43
+ > CREATIVE COMMONS CORPORATION IS NOT A LAW FIRM AND DOES NOT PROVIDE
44
+ > LEGAL SERVICES. DISTRIBUTION OF THIS DOCUMENT DOES NOT CREATE AN
45
+ > ATTORNEY-CLIENT RELATIONSHIP. CREATIVE COMMONS PROVIDES THIS
46
+ > INFORMATION ON AN "AS-IS" BASIS. CREATIVE COMMONS MAKES NO WARRANTIES
47
+ > REGARDING THE USE OF THIS DOCUMENT OR THE INFORMATION OR WORKS
48
+ > PROVIDED HEREUNDER, AND DISCLAIMS LIABILITY FOR DAMAGES RESULTING FROM
49
+ > THE USE OF THIS DOCUMENT OR THE INFORMATION OR WORKS PROVIDED
50
+ > HEREUNDER.
51
+ >
52
+ > Statement of Purpose
53
+ >
54
+ > The laws of most jurisdictions throughout the world automatically confer
55
+ > exclusive Copyright and Related Rights (defined below) upon the creator
56
+ > and subsequent owner(s) (each and all, an "owner") of an original work of
57
+ > authorship and/or a database (each, a "Work").
58
+ >
59
+ > Certain owners wish to permanently relinquish those rights to a Work for
60
+ > the purpose of contributing to a commons of creative, cultural and
61
+ > scientific works ("Commons") that the public can reliably and without fear
62
+ > of later claims of infringement build upon, modify, incorporate in other
63
+ > works, reuse and redistribute as freely as possible in any form whatsoever
64
+ > and for any purposes, including without limitation commercial purposes.
65
+ > These owners may contribute to the Commons to promote the ideal of a free
66
+ > culture and the further production of creative, cultural and scientific
67
+ > works, or to gain reputation or greater distribution for their Work in
68
+ > part through the use and efforts of others.
69
+ >
70
+ > For these and/or other purposes and motivations, and without any
71
+ > expectation of additional consideration or compensation, the person
72
+ > associating CC0 with a Work (the "Affirmer"), to the extent that he or she
73
+ > is an owner of Copyright and Related Rights in the Work, voluntarily
74
+ > elects to apply CC0 to the Work and publicly distribute the Work under its
75
+ > terms, with knowledge of his or her Copyright and Related Rights in the
76
+ > Work and the meaning and intended legal effect of CC0 on those rights.
77
+ >
78
+ > 1. Copyright and Related Rights. A Work made available under CC0 may be
79
+ > protected by copyright and related or neighboring rights ("Copyright and
80
+ > Related Rights"). Copyright and Related Rights include, but are not
81
+ > limited to, the following:
82
+ >
83
+ > i. the right to reproduce, adapt, distribute, perform, display,
84
+ > communicate, and translate a Work;
85
+ > ii. moral rights retained by the original author(s) and/or performer(s);
86
+ > iii. publicity and privacy rights pertaining to a person's image or
87
+ > likeness depicted in a Work;
88
+ > iv. rights protecting against unfair competition in regards to a Work,
89
+ > subject to the limitations in paragraph 4(a), below;
90
+ > v. rights protecting the extraction, dissemination, use and reuse of data
91
+ > in a Work;
92
+ > vi. database rights (such as those arising under Directive 96/9/EC of the
93
+ > European Parliament and of the Council of 11 March 1996 on the legal
94
+ > protection of databases, and under any national implementation
95
+ > thereof, including any amended or successor version of such
96
+ > directive); and
97
+ > vii. other similar, equivalent or corresponding rights throughout the
98
+ > world based on applicable law or treaty, and any national
99
+ > implementations thereof.
100
+ >
101
+ > 2. Waiver. To the greatest extent permitted by, but not in contravention
102
+ > of, applicable law, Affirmer hereby overtly, fully, permanently,
103
+ > irrevocably and unconditionally waives, abandons, and surrenders all of
104
+ > Affirmer's Copyright and Related Rights and associated claims and causes
105
+ > of action, whether now known or unknown (including existing as well as
106
+ > future claims and causes of action), in the Work (i) in all territories
107
+ > worldwide, (ii) for the maximum duration provided by applicable law or
108
+ > treaty (including future time extensions), (iii) in any current or future
109
+ > medium and for any number of copies, and (iv) for any purpose whatsoever,
110
+ > including without limitation commercial, advertising or promotional
111
+ > purposes (the "Waiver"). Affirmer makes the Waiver for the benefit of each
112
+ > member of the public at large and to the detriment of Affirmer's heirs and
113
+ > successors, fully intending that such Waiver shall not be subject to
114
+ > revocation, rescission, cancellation, termination, or any other legal or
115
+ > equitable action to disrupt the quiet enjoyment of the Work by the public
116
+ > as contemplated by Affirmer's express Statement of Purpose.
117
+ >
118
+ > 3. Public License Fallback. Should any part of the Waiver for any reason
119
+ > be judged legally invalid or ineffective under applicable law, then the
120
+ > Waiver shall be preserved to the maximum extent permitted taking into
121
+ > account Affirmer's express Statement of Purpose. In addition, to the
122
+ > extent the Waiver is so judged Affirmer hereby grants to each affected
123
+ > person a royalty-free, non transferable, non sublicensable, non exclusive,
124
+ > irrevocable and unconditional license to exercise Affirmer's Copyright and
125
+ > Related Rights in the Work (i) in all territories worldwide, (ii) for the
126
+ > maximum duration provided by applicable law or treaty (including future
127
+ > time extensions), (iii) in any current or future medium and for any number
128
+ > of copies, and (iv) for any purpose whatsoever, including without
129
+ > limitation commercial, advertising or promotional purposes (the
130
+ > "License"). The License shall be deemed effective as of the date CC0 was
131
+ > applied by Affirmer to the Work. Should any part of the License for any
132
+ > reason be judged legally invalid or ineffective under applicable law, such
133
+ > partial invalidity or ineffectiveness shall not invalidate the remainder
134
+ > of the License, and in such case Affirmer hereby affirms that he or she
135
+ > will not (i) exercise any of his or her remaining Copyright and Related
136
+ > Rights in the Work or (ii) assert any associated claims and causes of
137
+ > action with respect to the Work, in either case contrary to Affirmer's
138
+ > express Statement of Purpose.
139
+ >
140
+ > 4. Limitations and Disclaimers.
141
+ >
142
+ > a. No trademark or patent rights held by Affirmer are waived, abandoned,
143
+ > surrendered, licensed or otherwise affected by this document.
144
+ > b. Affirmer offers the Work as-is and makes no representations or
145
+ > warranties of any kind concerning the Work, express, implied,
146
+ > statutory or otherwise, including without limitation warranties of
147
+ > title, merchantability, fitness for a particular purpose, non
148
+ > infringement, or the absence of latent or other defects, accuracy, or
149
+ > the present or absence of errors, whether or not discoverable, all to
150
+ > the greatest extent permissible under applicable law.
151
+ > c. Affirmer disclaims responsibility for clearing rights of other persons
152
+ > that may apply to the Work or any use thereof, including without
153
+ > limitation any person's Copyright and Related Rights in the Work.
154
+ > Further, Affirmer disclaims responsibility for obtaining any necessary
155
+ > consents, permissions or other rights required for any use of the
156
+ > Work.
157
+ > d. Affirmer understands and acknowledges that Creative Commons is not a
158
+ > party to this document and has no duty or obligation with respect to
159
+ > this CC0 or use of the Work.
160
+
161
+ ## Tokenizer Data
162
+
163
+ ### OpenAI tiktoken Encoding Data
164
+
165
+ **Asset Information:**
166
+ - **File(s)**: `o200k_base.tiktoken` (pre-cached in Docker image at `/opt/tiktoken_cache/`)
167
+ - **Source**: [openai/tiktoken](https://github.com/openai/tiktoken) via `https://openaipublic.blob.core.windows.net/encodings/o200k_base.tiktoken`
168
+ - **Author**: OpenAI and Shantanu Jain
169
+ - **License**: MIT License
170
+
171
+ **License Text:**
172
+
173
+ > MIT License
174
+ >
175
+ > Copyright (c) 2022 OpenAI, Shantanu Jain
176
+ >
177
+ > Permission is hereby granted, free of charge, to any person obtaining a copy
178
+ > of this software and associated documentation files (the "Software"), to deal
179
+ > in the Software without restriction, including without limitation the rights
180
+ > to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
181
+ > copies of the Software, and to permit persons to whom the Software is
182
+ > furnished to do so, subject to the following conditions:
183
+ >
184
+ > The above copyright notice and this permission notice shall be included in all
185
+ > copies or substantial portions of the Software.
186
+ >
187
+ > THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
188
+ > IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
189
+ > FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
190
+ > AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
191
+ > LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
192
+ > OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
193
+ > SOFTWARE.
194
+
195
+ ## Usage Summary
196
+
197
+ - **Public domain assets** (ShareGPT conversation data) - no legal restrictions on use
198
+ - **MIT-licensed assets** (tiktoken encoding data) - permissive, requires copyright notice retention
199
+ - **All assets compatible with Apache 2.0** license used by this project
200
+ - **No endorsement implied** by original sources
201
+
202
+ ---
203
+ *Last updated: September 15, 2025*
@@ -0,0 +1,201 @@
1
+ Apache License
2
+ Version 2.0, January 2004
3
+ http://www.apache.org/licenses/
4
+
5
+ TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
6
+
7
+ 1. Definitions.
8
+
9
+ "License" shall mean the terms and conditions for use, reproduction,
10
+ and distribution as defined by Sections 1 through 9 of this document.
11
+
12
+ "Licensor" shall mean the copyright owner or entity authorized by
13
+ the copyright owner that is granting the License.
14
+
15
+ "Legal Entity" shall mean the union of the acting entity and all
16
+ other entities that control, are controlled by, or are under common
17
+ control with that entity. For the purposes of this definition,
18
+ "control" means (i) the power, direct or indirect, to cause the
19
+ direction or management of such entity, whether by contract or
20
+ otherwise, or (ii) ownership of fifty percent (50%) or more of the
21
+ outstanding shares, or (iii) beneficial ownership of such entity.
22
+
23
+ "You" (or "Your") shall mean an individual or Legal Entity
24
+ exercising permissions granted by this License.
25
+
26
+ "Source" form shall mean the preferred form for making modifications,
27
+ including but not limited to software source code, documentation
28
+ source, and configuration files.
29
+
30
+ "Object" form shall mean any form resulting from mechanical
31
+ transformation or translation of a Source form, including but
32
+ not limited to compiled object code, generated documentation,
33
+ and conversions to other media types.
34
+
35
+ "Work" shall mean the work of authorship, whether in Source or
36
+ Object form, made available under the License, as indicated by a
37
+ copyright notice that is included in or attached to the work
38
+ (an example is provided in the Appendix below).
39
+
40
+ "Derivative Works" shall mean any work, whether in Source or Object
41
+ form, that is based on (or derived from) the Work and for which the
42
+ editorial revisions, annotations, elaborations, or other modifications
43
+ represent, as a whole, an original work of authorship. For the purposes
44
+ of this License, Derivative Works shall not include works that remain
45
+ separable from, or merely link (or bind by name) to the interfaces of,
46
+ the Work and Derivative Works thereof.
47
+
48
+ "Contribution" shall mean any work of authorship, including
49
+ the original version of the Work and any modifications or additions
50
+ to that Work or Derivative Works thereof, that is intentionally
51
+ submitted to Licensor for inclusion in the Work by the copyright owner
52
+ or by an individual or Legal Entity authorized to submit on behalf of
53
+ the copyright owner. For the purposes of this definition, "submitted"
54
+ means any form of electronic, verbal, or written communication sent
55
+ to the Licensor or its representatives, including but not limited to
56
+ communication on electronic mailing lists, source code control systems,
57
+ and issue tracking systems that are managed by, or on behalf of, the
58
+ Licensor for the purpose of discussing and improving the Work, but
59
+ excluding communication that is conspicuously marked or otherwise
60
+ designated in writing by the copyright owner as "Not a Contribution."
61
+
62
+ "Contributor" shall mean Licensor and any individual or Legal Entity
63
+ on behalf of whom a Contribution has been received by Licensor and
64
+ subsequently incorporated within the Work.
65
+
66
+ 2. Grant of Copyright License. Subject to the terms and conditions of
67
+ this License, each Contributor hereby grants to You a perpetual,
68
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
69
+ copyright license to reproduce, prepare Derivative Works of,
70
+ publicly display, publicly perform, sublicense, and distribute the
71
+ Work and such Derivative Works in Source or Object form.
72
+
73
+ 3. Grant of Patent License. Subject to the terms and conditions of
74
+ this License, each Contributor hereby grants to You a perpetual,
75
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
76
+ (except as stated in this section) patent license to make, have made,
77
+ use, offer to sell, sell, import, and otherwise transfer the Work,
78
+ where such license applies only to those patent claims licensable
79
+ by such Contributor that are necessarily infringed by their
80
+ Contribution(s) alone or by combination of their Contribution(s)
81
+ with the Work to which such Contribution(s) was submitted. If You
82
+ institute patent litigation against any entity (including a
83
+ cross-claim or counterclaim in a lawsuit) alleging that the Work
84
+ or a Contribution incorporated within the Work constitutes direct
85
+ or contributory patent infringement, then any patent licenses
86
+ granted to You under this License for that Work shall terminate
87
+ as of the date such litigation is filed.
88
+
89
+ 4. Redistribution. You may reproduce and distribute copies of the
90
+ Work or Derivative Works thereof in any medium, with or without
91
+ modifications, and in Source or Object form, provided that You
92
+ meet the following conditions:
93
+
94
+ (a) You must give any other recipients of the Work or
95
+ Derivative Works a copy of this License; and
96
+
97
+ (b) You must cause any modified files to carry prominent notices
98
+ stating that You changed the files; and
99
+
100
+ (c) You must retain, in the Source form of any Derivative Works
101
+ that You distribute, all copyright, patent, trademark, and
102
+ attribution notices from the Source form of the Work,
103
+ excluding those notices that do not pertain to any part of
104
+ the Derivative Works; and
105
+
106
+ (d) If the Work includes a "NOTICE" text file as part of its
107
+ distribution, then any Derivative Works that You distribute must
108
+ include a readable copy of the attribution notices contained
109
+ within such NOTICE file, excluding those notices that do not
110
+ pertain to any part of the Derivative Works, in at least one
111
+ of the following places: within a NOTICE text file distributed
112
+ as part of the Derivative Works; within the Source form or
113
+ documentation, if provided along with the Derivative Works; or,
114
+ within a display generated by the Derivative Works, if and
115
+ wherever such third-party notices normally appear. The contents
116
+ of the NOTICE file are for informational purposes only and
117
+ do not modify the License. You may add Your own attribution
118
+ notices within Derivative Works that You distribute, alongside
119
+ or as an addendum to the NOTICE text from the Work, provided
120
+ that such additional attribution notices cannot be construed
121
+ as modifying the License.
122
+
123
+ You may add Your own copyright statement to Your modifications and
124
+ may provide additional or different license terms and conditions
125
+ for use, reproduction, or distribution of Your modifications, or
126
+ for any such Derivative Works as a whole, provided Your use,
127
+ reproduction, and distribution of the Work otherwise complies with
128
+ the conditions stated in this License.
129
+
130
+ 5. Submission of Contributions. Unless You explicitly state otherwise,
131
+ any Contribution intentionally submitted for inclusion in the Work
132
+ by You to the Licensor shall be under the terms and conditions of
133
+ this License, without any additional terms or conditions.
134
+ Notwithstanding the above, nothing herein shall supersede or modify
135
+ the terms of any separate license agreement you may have executed
136
+ with Licensor regarding such Contributions.
137
+
138
+ 6. Trademarks. This License does not grant permission to use the trade
139
+ names, trademarks, service marks, or product names of the Licensor,
140
+ except as required for reasonable and customary use in describing the
141
+ origin of the Work and reproducing the content of the NOTICE file.
142
+
143
+ 7. Disclaimer of Warranty. Unless required by applicable law or
144
+ agreed to in writing, Licensor provides the Work (and each
145
+ Contributor provides its Contributions) on an "AS IS" BASIS,
146
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
147
+ implied, including, without limitation, any warranties or conditions
148
+ of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
149
+ PARTICULAR PURPOSE. You are solely responsible for determining the
150
+ appropriateness of using or redistributing the Work and assume any
151
+ risks associated with Your exercise of permissions under this License.
152
+
153
+ 8. Limitation of Liability. In no event and under no legal theory,
154
+ whether in tort (including negligence), contract, or otherwise,
155
+ unless required by applicable law (such as deliberate and grossly
156
+ negligent acts) or agreed to in writing, shall any Contributor be
157
+ liable to You for damages, including any direct, indirect, special,
158
+ incidental, or consequential damages of any character arising as a
159
+ result of this License or out of the use or inability to use the
160
+ Work (including but not limited to damages for loss of goodwill,
161
+ work stoppage, computer failure or malfunction, or any and all
162
+ other commercial damages or losses), even if such Contributor
163
+ has been advised of the possibility of such damages.
164
+
165
+ 9. Accepting Warranty or Additional Liability. While redistributing
166
+ the Work or Derivative Works thereof, You may choose to offer,
167
+ and charge a fee for, acceptance of support, warranty, indemnity,
168
+ or other liability obligations and/or rights consistent with this
169
+ License. However, in accepting such obligations, You may act only
170
+ on Your own behalf and on Your sole responsibility, not on behalf
171
+ of any other Contributor, and only if You agree to indemnify,
172
+ defend, and hold each Contributor harmless for any liability
173
+ incurred by, or claims asserted against, such Contributor by reason
174
+ of your accepting any such warranty or additional liability.
175
+
176
+ END OF TERMS AND CONDITIONS
177
+
178
+ APPENDIX: How to apply the Apache License to your work.
179
+
180
+ To apply the Apache License to your work, attach the following
181
+ boilerplate notice, with the fields enclosed by brackets "[]"
182
+ replaced with your own identifying information. (Don't include
183
+ the brackets!) The text should be enclosed in the appropriate
184
+ comment syntax for the file format. We also recommend that a
185
+ file or class name and description of purpose be included on the
186
+ same "printed page" as the copyright notice for easier
187
+ identification within third-party archives.
188
+
189
+ Copyright [yyyy] [name of copyright owner]
190
+
191
+ Licensed under the Apache License, Version 2.0 (the "License");
192
+ you may not use this file except in compliance with the License.
193
+ You may obtain a copy of the License at
194
+
195
+ http://www.apache.org/licenses/LICENSE-2.0
196
+
197
+ Unless required by applicable law or agreed to in writing, software
198
+ distributed under the License is distributed on an "AS IS" BASIS,
199
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
200
+ See the License for the specific language governing permissions and
201
+ limitations under the License.
@@ -0,0 +1,354 @@
1
+ Metadata-Version: 2.4
2
+ Name: aiperf
3
+ Version: 0.11.0.dev20260717
4
+ Summary: AIPerf is a package for performance testing of AI models
5
+ Project-URL: Repository, https://github.com/ai-dynamo/aiperf/commit/0ede19ad5d9461a204f2f39befec5faf9f7e6b08
6
+ Author-email: "NVIDIA Inc." <sw-dl-dynamo@nvidia.com>
7
+ License: Apache-2.0
8
+ License-File: ATTRIBUTIONS.md
9
+ License-File: LICENSE
10
+ Requires-Python: <3.14,>=3.11
11
+ Requires-Dist: aiofiles~=24.1.0
12
+ Requires-Dist: aiohttp~=3.13.3
13
+ Requires-Dist: crick~=0.0.8
14
+ Requires-Dist: cyclopts<5,>=4
15
+ Requires-Dist: dash-bootstrap-components~=2.0.0
16
+ Requires-Dist: dash~=3.1.0
17
+ Requires-Dist: datasets>=3.0; platform_system != 'Windows' or platform_machine != 'ARM64'
18
+ Requires-Dist: fastapi<1,>=0.115
19
+ Requires-Dist: ffmpeg-python~=0.2.0
20
+ Requires-Dist: huggingface-hub<2.0,>=0.34.0
21
+ Requires-Dist: jinja2~=3.1.5
22
+ Requires-Dist: jmespath~=1.0.1
23
+ Requires-Dist: kaleido~=1.2.0
24
+ Requires-Dist: matplotlib>=3.10.0
25
+ Requires-Dist: msgspec<1.0.0,>=0.19.0
26
+ Requires-Dist: numpy<3,>=1.26.4
27
+ Requires-Dist: nvidia-ml-py
28
+ Requires-Dist: optuna>=3.6
29
+ Requires-Dist: orjson~=3.11.7
30
+ Requires-Dist: pandas~=2.3.3
31
+ Requires-Dist: pillow~=12.2.0
32
+ Requires-Dist: plotly~=6.4.0
33
+ Requires-Dist: prometheus-client~=0.23.1
34
+ Requires-Dist: protobuf>=4
35
+ Requires-Dist: psutil~=7.0.0
36
+ Requires-Dist: pyarrow>=18.0.0; platform_system != 'Windows' or platform_machine != 'ARM64'
37
+ Requires-Dist: pydantic-settings<3.0.0,>=2.10.0
38
+ Requires-Dist: pydantic<3.0.0,>=2.10.0
39
+ Requires-Dist: pyzmq~=26.4.0
40
+ Requires-Dist: rich~=14.1.0
41
+ Requires-Dist: ruamel-yaml~=0.18.12
42
+ Requires-Dist: scipy>=1.13.0
43
+ Requires-Dist: seaborn~=0.13.2
44
+ Requires-Dist: sentencepiece>=0.2.0
45
+ Requires-Dist: setproctitle~=1.3.6
46
+ Requires-Dist: soundfile~=0.13.1
47
+ Requires-Dist: starlette-compress>=1.0.1
48
+ Requires-Dist: textual~=5.3.0
49
+ Requires-Dist: tiktoken<1,>=0.7.0
50
+ Requires-Dist: tqdm>=4.67.1
51
+ Requires-Dist: transformers>=4.56.0
52
+ Requires-Dist: uvicorn[standard]<1,>=0.34
53
+ Requires-Dist: uvloop>=0.22.1; platform_system != 'Windows'
54
+ Requires-Dist: zstandard>=0.25.0
55
+ Provides-Extra: accuracy
56
+ Requires-Dist: deepeval<5.0.0,>=2.9.0; extra == 'accuracy'
57
+ Requires-Dist: latex2sympy2-extended>=1.0.6; extra == 'accuracy'
58
+ Requires-Dist: lighteval>=0.13.0; extra == 'accuracy'
59
+ Requires-Dist: sympy>=1.14.0; extra == 'accuracy'
60
+ Provides-Extra: botorch
61
+ Requires-Dist: botorch>=0.10; extra == 'botorch'
62
+ Requires-Dist: gpytorch>=1.11; extra == 'botorch'
63
+ Requires-Dist: optuna-integration>=3.6; extra == 'botorch'
64
+ Requires-Dist: torch>=2.0; extra == 'botorch'
65
+ Provides-Extra: dev
66
+ Requires-Dist: black>=25.1.0; extra == 'dev'
67
+ Requires-Dist: httpx>=0.27.0; extra == 'dev'
68
+ Requires-Dist: hypothesis<6.156,>=6.0.0; (platform_system == 'Windows' and platform_machine == 'ARM64') and extra == 'dev'
69
+ Requires-Dist: hypothesis>=6.0.0; (platform_system != 'Windows' or platform_machine != 'ARM64') and extra == 'dev'
70
+ Requires-Dist: jsonschema>=4.0.0; extra == 'dev'
71
+ Requires-Dist: looptime>=0.5; extra == 'dev'
72
+ Requires-Dist: mlflow<4.0.0,>=3.10.0; extra == 'dev'
73
+ Requires-Dist: opentelemetry-exporter-otlp-proto-http<2.0.0,>=1.24.0; extra == 'dev'
74
+ Requires-Dist: opentelemetry-sdk<2.0.0,>=1.24.0; extra == 'dev'
75
+ Requires-Dist: pre-commit>=4.2.0; extra == 'dev'
76
+ Requires-Dist: pytest-asyncio; extra == 'dev'
77
+ Requires-Dist: pytest-cov; extra == 'dev'
78
+ Requires-Dist: pytest-rerunfailures>=15.0; extra == 'dev'
79
+ Requires-Dist: pytest-xdist>=3.8.0; extra == 'dev'
80
+ Requires-Dist: pytest>=7.0.0; extra == 'dev'
81
+ Requires-Dist: ruff<0.15.0,>=0.14.0; extra == 'dev'
82
+ Requires-Dist: trustme>=1.0.0; (platform_system != 'Windows' or platform_machine != 'ARM64') and extra == 'dev'
83
+ Requires-Dist: wandb<1.0.0,>=0.19.0; extra == 'dev'
84
+ Provides-Extra: mlflow
85
+ Requires-Dist: mlflow<4.0.0,>=3.10.0; extra == 'mlflow'
86
+ Provides-Extra: optuna
87
+ Requires-Dist: botorch>=0.10; extra == 'optuna'
88
+ Requires-Dist: gpytorch>=1.11; extra == 'optuna'
89
+ Requires-Dist: optuna-integration>=3.6; extra == 'optuna'
90
+ Requires-Dist: torch>=2.0; extra == 'optuna'
91
+ Provides-Extra: otel
92
+ Requires-Dist: opentelemetry-exporter-otlp-proto-http<2.0.0,>=1.24.0; extra == 'otel'
93
+ Requires-Dist: opentelemetry-sdk<2.0.0,>=1.24.0; extra == 'otel'
94
+ Provides-Extra: test
95
+ Requires-Dist: httpx>=0.27.0; extra == 'test'
96
+ Requires-Dist: hypothesis<6.156,>=6.0.0; (platform_system == 'Windows' and platform_machine == 'ARM64') and extra == 'test'
97
+ Requires-Dist: hypothesis>=6.0.0; (platform_system != 'Windows' or platform_machine != 'ARM64') and extra == 'test'
98
+ Requires-Dist: jsonschema>=4.0.0; extra == 'test'
99
+ Requires-Dist: looptime>=0.5; extra == 'test'
100
+ Requires-Dist: pytest-asyncio; extra == 'test'
101
+ Requires-Dist: pytest-cov; extra == 'test'
102
+ Requires-Dist: pytest-rerunfailures>=15.0; extra == 'test'
103
+ Requires-Dist: pytest-xdist>=3.8.0; extra == 'test'
104
+ Requires-Dist: pytest>=7.0.0; extra == 'test'
105
+ Requires-Dist: trustme>=1.0.0; (platform_system != 'Windows' or platform_machine != 'ARM64') and extra == 'test'
106
+ Provides-Extra: wandb
107
+ Requires-Dist: wandb<1.0.0,>=0.19.0; extra == 'wandb'
108
+ Description-Content-Type: text/markdown
109
+
110
+ <!--
111
+ SPDX-FileCopyrightText: Copyright (c) 2025-2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
112
+ SPDX-License-Identifier: Apache-2.0
113
+ -->
114
+
115
+ # AIPerf
116
+
117
+ [![PyPI version](https://img.shields.io/pypi/v/AIPerf)](https://pypi.org/project/aiperf/)
118
+ [![License](https://img.shields.io/badge/License-Apache_2.0-blue.svg)](https://opensource.org/licenses/Apache-2.0)
119
+ [![Codecov](https://codecov.io/gh/ai-dynamo/aiperf/graph/badge.svg)](https://codecov.io/gh/ai-dynamo/aiperf)
120
+ [![Discord](https://dcbadge.limes.pink/api/server/D92uqZRjCZ?style=flat)](https://discord.gg/D92uqZRjCZ)
121
+ [![Ask DeepWiki](https://deepwiki.com/badge.svg)](https://deepwiki.com/ai-dynamo/aiperf)
122
+
123
+ AIPerf is a comprehensive benchmarking tool that measures the performance of generative AI models served by your preferred inference solution. It provides detailed metrics using a command line display as well as extensive benchmark performance reports.
124
+
125
+ <img width="1724" height="670" alt="AIPerf UI Dashboard" src="https://github.com/user-attachments/assets/7eb40867-b1c1-4ebe-bd57-7619f2154bba" />
126
+
127
+ ## Quick Start
128
+
129
+ This quick start guide leverages [Ollama](https://ollama.com/) via
130
+ [Docker Desktop](https://www.docker.com/products/docker-desktop/).
131
+
132
+ ### Setting up a Local Server
133
+
134
+ In order to set up an Ollama server, run `granite4:350m` using the following commands:
135
+
136
+ ```bash
137
+ docker run -d \
138
+ --name ollama \
139
+ -p 11434:11434 \
140
+ -v ollama-data:/root/.ollama \
141
+ ollama/ollama:latest
142
+ docker exec -it ollama ollama pull granite4:350m
143
+ ```
144
+
145
+ ### Basic Usage
146
+
147
+ Create a virtual environment and install AIPerf:
148
+
149
+ ```bash
150
+ python3 -m venv venv
151
+ source venv/bin/activate
152
+ pip install aiperf
153
+ ```
154
+
155
+ > [!NOTE]
156
+ > On Linux **aarch64** (`arm64`), one of AIPerf's dependencies (`crick`)
157
+ > ships only an sdist and needs a C compiler at install time. Install
158
+ > the system build toolchain before `pip install aiperf` —
159
+ > `sudo apt install build-essential` (Debian/Ubuntu),
160
+ > `sudo yum groupinstall "Development Tools"` (RHEL/CentOS), or
161
+ > equivalent. Linux x86_64, macOS, and Windows install from pre-built
162
+ > wheels and need no toolchain.
163
+
164
+ Optional integrations:
165
+ - `pip install "aiperf[mlflow]"` enables MLflow uploads and live telemetry streaming
166
+ - `pip install "aiperf[otel]"` enables OpenTelemetry metric streaming
167
+ - `pip install "aiperf[wandb]"` enables Weights & Biases result uploads
168
+ - `pip install "aiperf[mlflow,otel,wandb]"` installs all telemetry extras
169
+
170
+ To run a simple benchmark against your Ollama server:
171
+
172
+ ```bash
173
+ aiperf profile \
174
+ --model "granite4:350m" \
175
+ --streaming \
176
+ --endpoint-type chat \
177
+ --tokenizer ibm-granite/granite-4.0-micro \
178
+ --url http://localhost:11434 \
179
+ --request-count 10
180
+ ```
181
+
182
+
183
+ ### Example with Custom Configuration
184
+
185
+ ```bash
186
+ aiperf profile \
187
+ --model "granite4:350m" \
188
+ --streaming \
189
+ --endpoint-type chat \
190
+ --tokenizer ibm-granite/granite-4.0-micro \
191
+ --url http://localhost:11434 \
192
+ --concurrency 5 \
193
+ --request-count 10
194
+ ```
195
+
196
+ Example output:
197
+
198
+
199
+ **NOTE:** The example performance is reflective of a CPU-only run and does not represent an official benchmark.
200
+
201
+ ```bash
202
+ NVIDIA AIPerf | LLM Metrics
203
+ ┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┳━━━━━━━━━━┳━━━━━━━━━━━┳━━━━━━━━━━━┳━━━━━━━━━━━┳━━━━━━━━━━━┳━━━━━━━━━━┓
204
+ ┃ Metric ┃ avg ┃ min ┃ max ┃ p99 ┃ p90 ┃ p50 ┃ std ┃
205
+ ┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━╇━━━━━━━━━━╇━━━━━━━━━━━╇━━━━━━━━━━━╇━━━━━━━━━━━╇━━━━━━━━━━━╇━━━━━━━━━━┩
206
+ │ Time to First Token (ms) │ 7,463.28 │ 7,125.81 │ 9,484.24 │ 9,295.48 │ 7,596.62 │ 7,240.23 │ 677.23 │
207
+ │ Time to Second Token (ms) │ 68.73 │ 32.01 │ 102.86 │ 102.55 │ 99.80 │ 67.37 │ 24.95 │
208
+ │ Time to First Output Token (ms) │ 7,463.28 │ 7,125.81 │ 9,484.24 │ 9,295.48 │ 7,596.62 │ 7,240.23 │ 677.23 │
209
+ │ Request Latency (ms) │ 13,829.40 │ 9,029.36 │ 27,905.46 │ 27,237.77 │ 21,228.48 │ 11,338.31 │ 5,614.32 │
210
+ │ Inter Token Latency (ms) │ 65.31 │ 53.06 │ 81.31 │ 81.24 │ 80.64 │ 63.79 │ 9.09 │
211
+ │ Output Token Throughput Per User │ 15.60 │ 12.30 │ 18.85 │ 18.77 │ 18.08 │ 15.68 │ 2.05 │
212
+ │ (tokens/sec/user) │ │ │ │ │ │ │ │
213
+ │ Output Sequence Length (tokens) │ 95.20 │ 29.00 │ 295.00 │ 283.12 │ 176.20 │ 63.00 │ 77.08 │
214
+ │ Input Sequence Length (tokens) │ 550.00 │ 550.00 │ 550.00 │ 550.00 │ 550.00 │ 550.00 │ 0.00 │
215
+ │ Output Token Throughput (tokens/sec) │ 6.85 │ N/A │ N/A │ N/A │ N/A │ N/A │ N/A │
216
+ │ Request Throughput (requests/sec) │ 0.07 │ N/A │ N/A │ N/A │ N/A │ N/A │ N/A │
217
+ │ Request Count (requests) │ 10.00 │ N/A │ N/A │ N/A │ N/A │ N/A │ N/A │
218
+ └──────────────────────────────────────┴───────────┴──────────┴───────────┴───────────┴───────────┴───────────┴──────────┘
219
+
220
+ CLI Command: aiperf profile --model 'granite4:350m' --streaming --endpoint-type 'chat' --tokenizer 'ibm-granite/granite-4.0-micro' --url 'http://localhost:11434'
221
+ Benchmark Duration: 138.89 sec
222
+ CSV Export: /home/user/aiperf/artifacts/granite4:350m-openai-chat-concurrency1/profile_export_aiperf.csv
223
+ JSON Export: /home/user/Code/aiperf/artifacts/granite4:350m-openai-chat-concurrency1/profile_export_aiperf.json
224
+ Log File: /home/user/Code/aiperf/artifacts/granite4:350m-openai-chat-concurrency1/logs/aiperf.log
225
+ ```
226
+
227
+ ## Features
228
+
229
+ - Scalable multiprocess architecture with 10 services communicating via ZMQ
230
+ - 3 UI modes: `dashboard` (real-time TUI), `simple` (progress bars), `none` (headless)
231
+ - Multiple benchmarking modes: concurrency, request-rate, [request-rate with max concurrency](docs/tutorials/request-rate-concurrency.md), [trace replay](docs/benchmark-modes/trace-replay.md)
232
+ - Extensible plugin system for endpoints, datasets, transports, and metrics
233
+ - [Public dataset support](docs/benchmark-datasets.md) including ShareGPT and custom formats
234
+
235
+ ## Supported APIs
236
+
237
+ - OpenAI chat completions, completions, embeddings, audio, images
238
+ - NIM embeddings, rankings
239
+
240
+ ## Tutorials and Feature Guides
241
+
242
+ ### Getting Started
243
+ - [Basic Tutorial](docs/tutorial.md) - Profile Qwen3-0.6B with vLLM
244
+ - [Comprehensive Benchmarking Guide](docs/comprehensive-llm-benchmarking.md) - 5 real-world use cases
245
+ - [Interactive Chat Sanity Checks](docs/tutorials/interactive-chat.md) - Talk to an endpoint with `aiperf chat` and see per-turn TTFT/TPS/ITL
246
+ - [YAML Configuration Files](docs/tutorials/yaml-config.md) - Drive AIPerf from a config file instead of CLI flags
247
+ - [Sampling Distributions in YAML Configs](docs/tutorials/yaml-distributions.md) - Fixed, Normal, Log-normal, Multimodal, and Empirical shapes for ISL/OSL/turns/etc.
248
+ - [User Interface](docs/tutorials/ui-types.md) - Dashboard, simple, or headless
249
+ - [Hugging Face TGI](docs/tutorials/huggingface-tgi.md) - Profile Hugging Face TGI models
250
+ - [OpenAI Text Endpoints](docs/tutorials/openai-text-endpoints.md) - Profile OpenAI-compatible text APIs
251
+
252
+ ### Load Control and Timing
253
+ - [Request Rate with Max Concurrency](docs/tutorials/request-rate-concurrency.md) - Dual request control
254
+ - [Arrival Patterns](docs/tutorials/arrival-patterns.md) - Constant, Poisson, gamma traffic
255
+ - [Prefill Concurrency](docs/tutorials/prefill-concurrency.md) - Memory-safe long-context benchmarking
256
+ - [Adaptive Scale](docs/tutorials/adaptive-scale.md) - Single-run SLA boundary discovery and sustain
257
+ - [Gradual Ramping](docs/tutorials/ramping.md) - Smooth ramp-up of concurrency and request rate
258
+ - [Warmup Phase](docs/tutorials/warmup.md) - Eliminate cold-start effects
259
+ - [User-Centric Timing](docs/tutorials/user-centric-timing.md) - Per-user rate limiting for KV cache benchmarking
260
+ - [Request Cancellation](docs/tutorials/request-cancellation.md) - Timeout and resilience testing
261
+ - [Multi-URL Load Balancing](docs/tutorials/multi-url-load-balancing.md) - Distribute across servers
262
+
263
+ ### Workloads and Data
264
+ - [Trace Benchmarking](docs/benchmark-modes/trace-replay.md) - Deterministic workload replay
265
+ - [Bailian Traces](docs/tutorials/bailian-trace.md) - Bailian production trace replay
266
+ - [Baseten Traces](docs/tutorials/baseten-trace.md) - Baseten Parquet production trace replay
267
+ - [BurstGPT Traces](docs/tutorials/burst-gpt-trace.md) - BurstGPT real-world bursty traffic trace replay
268
+ - [SageMaker Data Capture](docs/tutorials/sagemaker-data-capture.md) - Replay production traffic from SageMaker endpoints
269
+ - [Custom Prompt Benchmarking](docs/tutorials/custom-prompt-benchmarking.md) - Send exact prompts as-is
270
+ - [Custom Dataset](docs/tutorials/custom-dataset.md) - Custom dataset formats
271
+ - [Inline Datasets](docs/tutorials/inline-datasets.md) - Embed records directly in the YAML config (single_turn, multi_turn, multi-pool random_pool, traces)
272
+ - [ShareGPT Dataset](docs/tutorials/sharegpt.md) - Profile with ShareGPT dataset
273
+ - [AIMO Dataset](docs/tutorials/aimo.md) - Profile with AIMO math reasoning datasets (NuminaMath-TIR, NuminaMath-CoT, NuminaMath-1.5, AIME)
274
+ - [MMStar Dataset](docs/tutorials/mmstar.md) - Profile vision language models with MMStar visual QA benchmark
275
+ - [MMVU Dataset](docs/tutorials/mmvu.md) - Profile video language models with MMVU expert-level video understanding benchmark
276
+ - [VisionArena Dataset](docs/tutorials/vision-arena.md) - Profile with real-world vision conversations from Chatbot Arena
277
+ - [LLaVA-OneVision Dataset](docs/tutorials/llava-onevision.md) - Profile with diverse multimodal instruction-following data
278
+ - [SPEED-Bench Dataset](docs/tutorials/speed-bench.md) - Profile speculative decoding with SPEED-Bench
279
+ - [InstructCoder Dataset](docs/tutorials/instruct-coder.md) - Profile with InstructCoder code generation dataset
280
+ - [SpecBench Dataset](docs/tutorials/spec-bench.md) - Profile with SpecBench speculative decoding dataset
281
+ - [Blazedit Dataset](docs/tutorials/blazedit.md) - Profile with Blazedit code editing dataset
282
+ - [ASR Datasets](docs/tutorials/asr.md) - Profile ASR models with LibriSpeech, VoxPopuli, GigaSpeech, AMI, and SPGISpeech
283
+ - [Synthetic Dataset Generation](docs/tutorials/synthetic-dataset.md) - Generate synthetic datasets
284
+ - [Agentic Code Generator](docs/tutorials/agentic-code-generator.md) - Generate multi-turn coding-agent traces for KV cache benchmarking
285
+ - [Fixed Schedule](docs/tutorials/fixed-schedule.md) - Precise timestamp-based execution
286
+ - [Time-based Benchmarking](docs/tutorials/time-based-benchmarking.md) - Duration-based testing
287
+ - [Sequence Distributions](docs/tutorials/sequence-distributions.md) - Mixed ISL/OSL pairings
288
+ - [Prefix Synthesis](docs/tutorials/prefix-synthesis.md) - Prefix data synthesis for KV cache testing
289
+ - [Reproducibility](docs/reproducibility.md) - Deterministic datasets with `--random-seed`
290
+ - [Template Endpoint](docs/tutorials/template-endpoint.md) - Custom Jinja2 request templates
291
+ - [Multi-Turn Conversations](docs/tutorials/multi-turn.md) - Multi-turn conversation benchmarking
292
+ - [Raw Payload Replay](docs/tutorials/raw-payload-replay.md) - Verbatim JSONL payload replay (single file or directory)
293
+ - [Inputs JSON Replay](docs/tutorials/inputs-json-replay.md) - Verbatim multi-turn replay of AIPerf inputs.json artifacts
294
+ - [Local Tokenizer](docs/tutorials/local-tokenizer.md) - Use local tokenizers without HuggingFace
295
+
296
+ ### Endpoint Types
297
+ - [Embeddings](docs/tutorials/embeddings.md) - Profile embedding models
298
+ - [Rankings](docs/tutorials/rankings.md) - Profile ranking models
299
+ - [OpenAI Responses API](docs/tutorials/openai-responses.md) - Profile OpenAI Responses API endpoints
300
+ - [Audio](docs/tutorials/audio.md) - Profile audio language models
301
+ - [NIM Image Retrieval](docs/tutorials/nim-image-retrieval.md) - Profile NIM image retrieval models
302
+ - [Vision](docs/tutorials/vision.md) - Profile vision language models
303
+ - [Image Generation](docs/tutorials/image-generation.md) - Benchmark any OpenAI-compatible image generation API
304
+ - [SGLang Image Edit](docs/tutorials/sglang-image-edit.md) - Benchmark OpenAI-compatible image-to-image (image edit) endpoints
305
+ - [SGLang Video Generation](docs/tutorials/sglang-video-generation.md) - Video generation benchmarking
306
+ - [Synthetic Video](docs/tutorials/synthetic-video.md) - Synthetic video generation
307
+
308
+ ### Analysis and Monitoring
309
+ - [Timeslice Metrics](docs/tutorials/timeslices.md) - Per-timeslice performance analysis
310
+ - [Goodput](docs/tutorials/goodput.md) - SLO-based throughput measurement
311
+ - [Parameter Sweeps](docs/tutorials/sweeps.md) - YAML reference for grid/zip/scenarios sweeps + multi-run, with picker for choosing a sweep mode
312
+ - [Adaptive Search](docs/tutorials/adaptive-search.md) - Bayesian-optimization walkthrough (single-objective + multi-objective Pareto)
313
+ - [Adaptive Scale](docs/tutorials/adaptive-scale.md) - Single-run SLA boundary discovery and sustain
314
+ - [Search Recipes](docs/sweeping/search-recipes.md) - Named recipe catalog including `pareto-sweep`, `max-throughput-ttft-sla`, `max-concurrency-under-sla`
315
+ - [HTTP Trace Metrics](docs/tutorials/http-trace-metrics.md) - DNS, TCP/TLS, TTFB timing
316
+ - [Multi-Run Confidence](docs/tutorials/multi-run-confidence.md) - Confidence intervals across repeated runs
317
+ - [Profile Exports](docs/tutorials/working-with-profile-exports.md) - Post-processing with Pydantic models
318
+ - [Visualization and Plotting](docs/tutorials/plot.md) - PNG charts and multi-run comparison
319
+ - [Auto-Plot After Profile](docs/tutorials/auto-plot.md) - Run `aiperf plot` automatically after `aiperf profile`
320
+ - [GPU Telemetry](docs/tutorials/gpu-telemetry.md) - DCGM metrics collection
321
+ - [OTel + MLflow Live Telemetry](docs/tutorials/otel-mlflow.md) - Stream metrics to OTel and MLflow in real time
322
+ - [Weights & Biases Export](docs/tutorials/wandb.md) - Upload results tables and artifacts to wandb
323
+ - [Server Metrics](docs/server-metrics/server-metrics.md) - Prometheus-compatible metrics
324
+
325
+ ## Documentation
326
+
327
+ | Document | Purpose |
328
+ |----------|---------|
329
+ | [Architecture](docs/architecture.md) | Three-plane architecture, core components, credit system, data flow |
330
+ | [CLI Options](docs/cli-options.md) | Complete command and option reference |
331
+ | [Metrics Reference](docs/metrics-reference.md) | All metric definitions, formulas, and requirements |
332
+ | [Environment Variables](docs/environment-variables.md) | All `AIPERF_*` configuration variables |
333
+ | [Plugin System](docs/plugins/plugin-system.md) | Plugin architecture, 25+ categories, creation guide |
334
+ | [Creating Plugins](docs/plugins/creating-your-first-plugin.md) | Step-by-step plugin tutorial |
335
+ | [Accuracy Benchmarks](docs/accuracy/accuracy-benchmarking.md) | Accuracy evaluation against MMLU, AIME, and other benchmarks |
336
+ | [Benchmark Modes](docs/benchmark-modes/trace-replay.md) | Trace replay and timing modes |
337
+ | [Server Metrics](docs/server-metrics/server-metrics.md) | Prometheus-compatible server metrics collection |
338
+ | [Tokenizer Auto-Detection](docs/reference/tokenizer-auto-detection.md) | Pre-flight tokenizer detection |
339
+ | [Conversation Context Mode](docs/reference/conversation-context-mode.md) | How conversation history accumulates in multi-turn |
340
+ | [Dataset Synthesis API](docs/api/synthesis.md) | Synthesis module API reference |
341
+ | [Code Patterns](docs/dev/patterns.md) | Code examples for services, models, messages, plugins |
342
+ | [Migrating from Genai-Perf](docs/migrating.md) | Migration guide and feature comparison |
343
+ | [Design Proposals](https://github.com/ai-dynamo/enhancements) | Enhancement proposals and discussions |
344
+
345
+ ## Contributing
346
+
347
+ See [CONTRIBUTING.md](CONTRIBUTING.md) for development setup, coding conventions, and contribution guidelines.
348
+
349
+ ## Known Issues
350
+
351
+ - Output sequence length constraints (`--output-tokens-mean`) cannot be guaranteed unless you pass `ignore_eos` and/or `min_tokens` via `--extra-inputs` to an inference server that supports them.
352
+ - Very high concurrency settings (typically >15,000) may lead to port exhaustion on some systems. Adjust system limits or reduce concurrency if connection failures occur.
353
+ - Startup errors caused by invalid configuration settings can cause AIPerf to hang indefinitely. Terminate the process and check configuration settings.
354
+ - Copying selected text may not work reliably in the dashboard UI. Use the `c` key to copy all logs.
@@ -0,0 +1,10 @@
1
+ # SPDX-FileCopyrightText: Copyright (c) 2025 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
2
+ # SPDX-License-Identifier: Apache-2
3
+
4
+ [build-system]
5
+ requires = ["wheel-stub"]
6
+ build-backend = "wheel_stub.buildapi"
7
+
8
+ [tool.wheel_stub]
9
+ index_url = "https://pypi.nvidia.com/"
10
+ include_cuda_debuginfo = true