Home
Forums
New posts
Search forums
What's new
New posts
New profile posts
Latest activity
Members
Current visitors
New profile posts
Search profile posts
Log in
Register
What's new
Search
Search
Search titles only
By:
New posts
New profile posts
Latest activity
Menu
Log in
Register
Install the app
Install
Home
What's new
JavaScript is disabled. For a better experience, please enable JavaScript in your browser before proceeding.
You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an
alternative browser
.
Latest activity
What's new
New posts
New profile posts
Latest activity
Andrey Ignatov
replied to the thread
Does aibenchmark update and adapt to AndroidT?
.
Hi @Mountain, replied to you by email.
Aug 6, 2023
Andrey Ignatov
posted the thread
AI Benchmark 5.1.0 Mobile Released
in
News, Releases and More
.
What's New: Added new NPU power consumption test. Updated TFLite runtime. Updated TFLite GPU, NNAPI, Qualcomm QNN, Hexagon NN and...
Aug 6, 2023
Emperor
reacted to
Mako443's post
in the thread
How to reproduce Float16 models on HTP
with
Love
.
Happy to help! And yes of course, let's connect. I'll send a DM. Cheers, Manuel
Jul 20, 2023
M
Mako443
replied to the thread
How to reproduce Float16 models on HTP
.
Happy to help! And yes of course, let's connect. I'll send a DM. Cheers, Manuel
Jul 19, 2023
M
Mountain
posted the thread
Does aibenchmark update and adapt to AndroidT?
in
General AI Benchmark Questions
.
hi, The current AndroidT version and release, there is a new nnapi operator, will aibenchmark update and adapt to this?
Jul 19, 2023
M
mr-robot
reacted to
Andrey Ignatov's post
in the thread
Difference between the HTP and DSP delegate?
with
Like
.
This is a very brief answer, but the general idea is as follows: HTP = rebranded compute DSP (since Snapdragon 888 / Hexagon v68)...
Jul 11, 2023
Emperor
reacted to
Andrey Ignatov's post
in the thread
Difference between the HTP and DSP delegate?
with
Like
.
This is a very brief answer, but the general idea is as follows: HTP = rebranded compute DSP (since Snapdragon 888 / Hexagon v68)...
Jul 10, 2023
Andrey Ignatov
replied to the thread
Difference between the HTP and DSP delegate?
.
This is a very brief answer, but the general idea is as follows: HTP = rebranded compute DSP (since Snapdragon 888 / Hexagon v68)...
Jul 10, 2023
Emperor
reacted to
mr-robot's post
in the thread
Difference between the HTP and DSP delegate?
with
Like
.
Thanks for the quick reply. Are you familiar with the internal components of the HTP? Eg the difference between the HTA, HVX and HMX...
Jul 10, 2023
M
mr-robot
replied to the thread
Difference between the HTP and DSP delegate?
.
Thanks for the quick reply. Are you familiar with the internal components of the HTP? Eg the difference between the HTA, HVX and HMX...
Jul 10, 2023
Andrey Ignatov
reacted to
Collin80's post
in the thread
Device List
with
Like
.
I recently got an S23 Ultra and ran the benchmark. The score was 2031K which does blow away anything else on the device list so far. You...
Jul 10, 2023
Emperor
reacted to
Andrey Ignatov's post
in the thread
Is it possible to get source code of AI bench 5.0.3
with
Like
.
One can potentially extract all models directly from the benchmark APK file. Feel free to use this forum for sharing or comparing the...
Jul 10, 2023
Emperor
reacted to
Andrey Ignatov's post
in the thread
Difference between the HTP and DSP delegate?
with
Like
.
Yes, partly: the Hexagon 6xx family is denoted as DSPs in QNN, while the Hexagon 7xx family - as HTPs. Here you can find the full list...
Jul 10, 2023
Emperor
reacted to
Andrey Ignatov's post
in the thread
Questions about Apple Benchmarks
with
Like
.
INT8 models were running with the TFLite GPU delegate. No, these are plain NPU/GPU runtime results. Because of the bug in the iOS...
Jul 10, 2023
Emperor
reacted to
Andrey Ignatov's post
in the thread
Power Efficiency Measurements & Inference Precision
with
Like
.
In the standard benchmark mode, only INT8 inference is tested. However, one can also check the results of FP16 inference in the PRO...
Jul 10, 2023
Home
What's new
Top