-
Hi, I want to use `iree-module-run` to inference huge model, but I encountered an "Argument list too long" error.
I try to save runtime status in the disk to avoid long `--input=...` option, but the …
-
### Description of the problem / feature request:
Bazel doesn't support cc_embed_data
### Feature requests: what underlying problem are you trying to solve with this feature?
I would like to …
-
This is an umbrella issue for allowing fp8 type(s) in shark, spanning all the required layers of the stack: Turbine, IREE, MLIR, LLVM, including backends of interest like ROCm.
Some initial researc…
kuhar updated
9 months ago
-
This EPIC tracks all the related work.
### Tasks related to core functionality
### Tasks
- [x] https://github.com/openxla/iree/issues/15680
- [x] https://github.com/openxla/iree/issues/15214…
-
This is the EPIC that collects pack optimization work. We should generate vectorized code (without stack allocation) after finishing below tasks.
### Tasks
- [ ] https://github.com/iree-org/ir…
-
### What happened?
Hi, I'm new to IREE, and I wanna build from source following documents [here](https://iree.dev/building-from-source/getting-started/#running-tests).
However, when I try to buil…
-
One of the issues faced during SDXL support (https://github.com/openxla/iree/pull/16854) was the missing support for operations added in LinalgExt on all codegen backends i.e, CPU, SPIRV and LLVMGPU.
…
-
### Request description
[ml_program](https://discourse.llvm.org/t/rfc-introduce-ml-program-dialect-and-top-level-ops-proposal-v2/60907/72) is being used for certain JAX -> IREE flows and we are looki…
-
### What happened?
The same .vmfb gives different results on ROCM and HIP hal drivers. Caching allocator is being used on both, but this doesn't seem to make a difference if disabled.
Good numer…
-
Currently I see some documentation on how to run SD from command line and
there are some docs about IREE profiling and blender plugin but I dont see documentation on how to build custom SHARK (curre…