f6f734f44d
vulkan int8 kernels ( #6751 )
...
* vulkan: add int8 kernels for core layers
Implement Vulkan int8 paths for convolution, convolutiondepthwise, innerproduct, and gemm with scalar fallback and integer dot product shader branches. Add goal.md and enable existing int8 unit tests for Vulkan coverage.
* enable int8 gpu benchmark
2026-06-17 17:39:49 +08:00
619ea8e0b2
change llm benchmark seqlen from 1024 to 256 ( #6738 )
2026-05-22 17:04:32 +08:00
244572aeb8
fix simplemath build, trigger ci on benchmark changes ( #6722 )
2026-05-14 11:27:58 +08:00
13bca30c00
benchncnn_llm ( #6711 )
2026-05-13 11:38:29 +08:00
155ab53695
move model files into sub directory ( #6710 )
2026-05-09 16:32:02 +08:00
d4f599f2cd
benchmark: use Gemm for vision transformer linear layers ( #6709 )
2026-05-09 14:29:54 +08:00
f4e9a78291
Add benchmark results for several instances from Microsoft Azure. ( #6552 )
2026-03-30 10:21:46 +08:00
c8e774f247
add Qualcomm Snapdragon X Elite benchmark ( #6535 )
2026-02-07 21:56:36 +08:00
6b4f6f95e5
[feat](benchmark): replace load_param with load_param_mem ( #6286 )
...
* [feat](benchmark): replace load_param with load_param_mem
* [docs](benchmark): update benchmark docs
* refactor(benchmark): remove model registry and optimize param loading
* fix(benchmark): correct README param description
* refactor(benchmark): rename header file
* update doc
---------
Co-authored-by: nihui <nihuini@tencent.com >
Co-authored-by: nihui <shuizhuyuanluo@126.com >
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com >
2026-01-08 14:23:00 +08:00
d0afafe7bd
Update README with Arduino UNO Q benchmark results ( #6454 )
2025-12-15 10:14:10 +08:00
b8a1c92cce
fix format specifier for size_t to support 64-bit values ( #6315 )
...
---------
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com >
2025-09-12 14:00:27 +08:00
b5bc05e2e8
drop vulkan shader pack8 ( #6270 )
2025-08-22 11:05:59 +08:00
4d8220ce22
Skip int8 model on GPU and merge from upstream ( #6174 )
2025-07-12 17:06:46 +08:00
171b9d1bba
use spdx license header, copyright Tencent ( #6152 )
2025-07-08 11:01:29 +08:00
24a3b99f1f
drop layer support_image_storage and option use_image_storage ( #6126 )
...
* fix pyncnn build
2025-06-16 11:23:41 +08:00
46b3de348c
Add MUSE Pi Pro Spacemit M1 benchmark. ( #6074 )
2025-05-26 22:35:55 +08:00
edfb2c6bee
Add MacBook Air (13-inch, M3, 2024) benchmark. ( #5645 )
2025-04-18 10:27:11 +08:00
55e5c0ba05
benchncnn: add radxa orion o6 benchmark ( #5943 )
2025-03-25 07:56:53 +08:00
cdcbb3b836
benchncnn: add vim3 vulkan benchmark ( #5821 )
2024-12-14 17:34:00 +08:00
07196eee2e
benchmark: add Kunpeng 920 7260 ( #5606 )
2024-08-16 10:50:26 +08:00
74037b49f8
Add Axera AX630C benchmark ( #5559 )
2024-07-08 12:59:13 +08:00
1225793e56
benchmark: add Snapdragon 765G and CVITEK SG2000 ( #5555 )
2024-07-08 11:07:11 +08:00
9f62738e7c
Add RaspberryPi 5 CPU Overclock benchmark. ( #5561 )
2024-07-08 11:01:12 +08:00
317635b2fb
add OrangePicm4 ( #5551 )
2024-07-02 12:37:53 +08:00
e48c8aa21c
benchmark: add OrangePi 5 Plus ( #5550 )
2024-07-01 10:46:18 +08:00
5d4f0213ce
Add Snapdragon 888 benchmark. ( #5548 )
2024-07-01 10:44:58 +08:00
c8e6e18918
benchmark: add EPYC 7742 and V100 ( #5528 )
2024-06-23 23:53:16 +08:00
b786af56f8
benchmark: add raspberry pi 5 benchmark after GPU overclock ( #5518 )
2024-06-19 15:52:22 +08:00
e389a15846
update qcom855plus benchmark
2024-06-12 15:08:15 +08:00
5fdbb54b7f
Add Radxa Zero 3W to benchmark ( #5321 )
2024-01-30 14:15:33 +08:00
10fd242741
add remipi benchmark ( #5319 )
2024-01-29 17:01:40 +08:00
66b26b6ce8
add PhytiumPi ncnn benchmark ( #5312 )
2024-01-26 13:08:27 +08:00
09f15e6980
Add Dimensity 9300 MT6989 ncnn benchmark ( #5284 )
...
Signed-off-by: Molly Sophia <mollysophia379@gmail.com >
2024-01-15 10:45:19 +08:00
28078b7ad6
add orangepi zero2 ncnn benchmark ( #5277 )
2024-01-12 15:30:59 +08:00
2ac1e776dd
Add Xeon Phi 3120A results ( #5276 )
2024-01-12 15:30:36 +08:00
237f45ff3e
Update AWS and OneCloud benchmark results with ncnn 20240102 ( #5257 )
...
* Update OneCloud results with 2024010220240102
2024-01-08 11:01:42 +08:00
ded0b78bb2
fix nvidia vulkan crash on exit ( #5234 )
2023-12-20 11:32:53 +08:00
9023ecf52c
update benchmark for visionfive2 ( #5217 )
2023-12-15 15:13:08 +08:00
070dac9955
Benchmark results Jetson Orin Nano, RPi 5 (refreshed), and a simple ranking tool ( #5216 )
2023-12-12 11:15:17 +08:00
3fcb8a7263
Add 2K2000 benchmark ( #5214 )
2023-12-11 11:45:54 +08:00
3cff53c30a
add Jetson TX2 benchmark ( #5213 )
2023-12-11 10:39:29 +08:00
5c432d5985
add the benchmark of 8gen2 ( #5180 )
2023-11-28 10:29:59 +08:00
a029587725
Update RPi 5 GPU Benchmark Result ( #5160 )
...
use mesa-23.3-rc4 driver
2023-11-21 14:15:57 +08:00
22c990ac65
Add Raspberry Pi 5 GPU Benchmark ( #5152 )
2023-11-15 17:09:04 +08:00
465debe9bb
Add print statements for 4 dimensions benchmark ( #5148 )
2023-11-15 11:43:59 +08:00
c96bc08307
Add OneCloud Amlogic S805 ( #5120 )
2023-10-31 17:34:34 +08:00
a1e3ebf8e5
implement simplemath ( #4905 )
...
* complete abs, fmod and sin function in simplemath.h
* remove some unused variables in simplemath.cpp
* modify test-coverage.yml and add some functions to simplemath.cpp
* modify erf.cpp which included math.h
* include platform.h for NCNN_SIMPLEMATH definition
* move utility constants and functions in simplemath.h to simplemath.cpp
* guard simplemath functions with extern "C"
* add NCNN_EXPORT macro in simplemath.h
* include plateform.h and guard all declarations with NCNN_SIMPLEMATH
* clean unused code in test_unaryop.cpp
* guard #include <vector> with NCNN_SIMPLEMATH in benchncnn.cpp
* add 'static' to guard functions that not declarated in header file
* modify sin and cos with better implementation
---------
Co-authored-by: HonestDeng <HonestDeng@users.noreply.github.com >
2023-10-23 10:51:51 +08:00
84aaedbe96
Added 5 devices(AMD*2, Intel*2, NVIDIA*1) benchmark result. ( #5085 )
2023-10-20 15:46:51 +08:00
d1289fb12d
benchmark: add RTX A3000 6G benchmark ( #5070 )
2023-10-10 16:36:55 +08:00
f38fc95e0c
benchmark: add raspberry pi 5 8G benchmark ( #5058 )
2023-09-28 14:45:40 +08:00
nihui