mlir

PyTorch's blog details the open-source MLIR-based compiler backend (KTIR to DataflowIR to SpyreCode) that translates PyTorch Inductor's decisions into executable dataflow pipelines for IBM's Spyre AI accelerator.

HardenedLinux 发布开源编译器优先的 AI 推理运行时 SonicBoom v0.0.3，声称在 RTX 3050 上解码速度超过 llama.cpp，并呼吁社区贡献 GPU 基准测试结果。

Modular 团队将参加 2026 年 LLVM 开发者大会，Mojo 团队举办全天工作坊讲解 Mojo 内部原理与路线图，Spenser Bauman 将分享 MAX 的 MLIR/LLVM 图编译器技术，并与 Qualcomm 联合赞助及现场招聘。

At PyTorch Conference North America 2026, AMD's Liz Li and Jiahui Cao will present "Extending TorchInductor with FlyDSL," an MLIR-native GPU kernel backend for high-performance GEMMs that integrates with TorchInductor's compilation and autotuning pipeline, with benchmark comparisons against Triton on AMD Instinct GPUs.

This article discusses how MLIR compiler passes can introduce NaN values through constant folding even with valid operands, and proposes a verification workflow to catch and fix such issues at the IR boundary.

This post explains MLIR as a compiler infrastructure framework that provides a flexible dialect stack, describing how it is used across modern ML compilers like XLA, Triton, and Mojo to progressively lower tensor operations to machine code.

This paper introduces benchmarks for natural language to MLIR code generation across multiple dialects and a schema-derived constrained decoding stack that allows a small language model to match or exceed large code LMs on structural verifier tasks, without retraining.

This paper presents an MLIR-based compilation method for large language models, using two custom dialects (TopOp and TpuOp) to lower models from framework-agnostic semantics to hardware-specific instructions. It also introduces a three-stage static compilation for autoregressive inference stages: prefill, prefill_kv, and decode.