-
The StableHLO Python bindings have been very handy.
This means I can run for example:
```python
from mlir.dialects import stablehlo
from mlir.ir import Context, Module
# Define the MLIR str…
-
we actually have had this working and it was still working up till last week [stateless_llama.py](https://github.com/nod-ai/SHARK-ModelDev/blob/main/models/turbine_models/custom_models/stateless_llama…
-
Step1: Change this [line](https://github.com/nod-ai/iree-amd-aie/blob/c15a45a016a504f032b0f5c02d64c9916da91521/build_tools/build_test_cpp.ps1#L61) from
"-DCMAKE_BUILD_TYPE=Release"
to
"-DCMAKE_BUIL…
-
Windows builds of shortfin have been failing since https://github.com/nod-ai/SHARK-Platform/pull/434
Logs just before: https://github.com/nod-ai/SHARK-Platform/actions/runs/11746567811/job/32726612…
-
This is an uber-issue for making the tuner easier to maintain. The current implementation has a few issues that make the tuner library fragile and prone to getting out of sync with the IREE compiler. …
kuhar updated
6 hours ago
-
For reproduction.
Input Model:
https://sharkpublic.blob.core.windows.net/sharkpublic/sai/sdxl-punet/punet.mlir
Input data :
wget https://sharkpublic.blob.core.windows.net/sharkpublic/sai/sdx…
-
It looks like the `all_gpu_id_array` is not cleaned up when KMT is unloaded. If KMT is initialized multiple times in the same process it will leak the array multiple times. `hsakmt_fmm_destroy_process…
-
# Description
There appears to be an accuracy drop in using shortfin LLM server with the latest iree-compiler, which causes failures in the [CPU LLM Server Integration Test](https://github.com/nod-…
-
### Describe the proposed feature and its relevance to inferencing
Add IREE and Godot Engine demos
### Context or use case
How can we should the world IREE and Godot Engine is cool.
### Pr…
-
For the attached IR, seeing crash as
```
iree-compile: /proj/xhdhdstaff6/dhirajp/localBuild/iree/third_party/llvm-project/mlir/lib/Transforms/Utils/DialectConversion.cpp:2868: SmallVector mlir::Ty…