Update blog

Author: cmdr2

content/blog/2025-10-10-1760088945.md

@@ -12,13 +12,13 @@ Some notes on machine-learning compilers, gathered while researching tech for Ea
 
 ## tl;dr summary
 
-The current state is (could change in the future!):
+The current state is:
 1. Vendor-specific compilers are the only performant options on consumer GPUs. For e.g. [TensorRT-RTX](https://docs.nvidia.com/deeplearning/tensorrt-rtx/latest/index.html) for NVIDIA, [MiGraphX](https://rocm.docs.amd.com/projects/AMDMIGraphX/en/latest/) for AMD, [OpenVINO](https://github.com/openvinotoolkit/openvino) for Intel.
 2. Cross-vendor compilers are just not performant enough for Stable Diffusion-class workloads on consumer GPUs. For e.g. like [TVM](https://tvm.apache.org/), [IREE](https://iree.dev/), [XLA](https://openxla.org/xla).
 
 The focus of cross-vendor compilers seems to be either on datacenter hardware, or embedded devices. The performance on desktops and laptops is pretty poor. Mojo doesn't target this category (and doesn't support Windows). Probably because datacenters and embedded devices are currently where the attention (and money) is.
 
-The idea of a cross-vendor ML compiler is clearly awesome, and I think this is the way things should go. But we're not there yet for desktops/laptops, in terms of runtime performance.
+This could change in the future! The idea of a cross-vendor ML compiler is clearly awesome, and I think this is the way things should go. But we're not there yet for desktops/laptops, in terms of runtime performance.
 
 ## What's an ML compiler?