Comments (4)
@wangyems, In MoeGemmRunner::dispatch_to_arch, I saw it dispatch for SM from 70 to 89. We shall skip MOE tests for other GPUs (< 70, and >= 90).
BTW, could you test MOE in V100 to see whether it could run in SM=70? I saw some feature required SM>=80:
I think the requirements are like:
float: sm 70 to 89 (In theory, it shall support all GPUs, just a limitation of MoeGemmRunner::dispatch_to_arch)
float16: sm 70 to 89
bfloat16: sm 80 to 89
from onnxruntime.
I tested it on V100 on Linux with CUDA 11.8.
[==========] 4 tests from 1 test suite ran. (10758 ms total)
[ PASSED ] 3 tests.
[ FAILED ] 1 test, listed below:
[ FAILED ] MoETest.QMoETest_Mixtral_Int4
from onnxruntime.
Our Linux training GPU machine pools use V100. Why they didn't catch this error?
from onnxruntime.
checking..
from onnxruntime.
Related Issues (20)
- CUDA wasnt able to be loaded when tring to use CUDA12.4 and onnxruntime-gpu 11.8 HOT 2
- Flan-T5 small converted model produces wrong result with batch size > 1 and long senetences
- [Training] Support for RKNPU Execution Provider on RK3562 Platform and On-Device Training Capabilities HOT 1
- Support Numpy v2.0 HOT 9
- Segmentation fault during inference HOT 2
- Can't find definition of OrtTensorRTProviderOptionsV2 in latest release
- [Mobile] Segmentation fault after repeated inference HOT 5
- [Bug] W16A16 quantization, qdq_error is empty, but W8A8 is normal HOT 1
- [Feature Request] support FP8 calibraion method and quantization HOT 2
- DML cannot use device_id = 1 , run_with_iobinding failed. HOT 1
- [Mobile] CreatMemoryInfo Failure in Android when try to create Ort::Value HOT 6
- [Build] ERROR: *.whl is not a valid wheel filename & ImportError: DLL load failed while importing onnxruntime_pybind11_state HOT 6
- [Performance] Running YOLOv8-seg.onnx with Dynamic Batch Size on GPU HOT 1
- [Web] `Error: [WebGPU] Kernel "[Conv] /text_encoder/encoder/layers.0/feed_forward/conv_2/Conv" failed. Error: FILTER_IN_CHANNEL should be equal to DATA_CHANNEL` HOT 1
- [Bug] fp8 quantization, weight_error is empty, but qdq_err is normal
- cannot resolve operator 'Cast' with opsets: ai.onnx v17 [Web]
- [Performance] Call a error when running on windows 7.detail:System.TypeInitializationException at Microsoft.ML.OnnxRuntime.SessionOptions..ctor() HOT 3
- [Feature Request] Support for Florence-2 model family HOT 1
- [Web] [Webnn] Can't seem to use the WebNN. Getting "WEBNN execution provider is not supported in this build." HOT 2
- Symbolic Shape infer fails on onnx file without much logs
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from onnxruntime.