Pull Requests onnx/onnx-mlir

Enable One-Shot Bufferization for Mixed Linalg and ONNX Operations

#3358 opened 2026-01-08 06:53 by kimm240

Implement Selective Linalg Conversion with --linalg-ops Option

#3356 opened 2026-01-07 05:31 by kimm240

[WIP] Handle onnx.Concat whose some inputs are tensor<0xdtype>

#3353 opened 2025-12-17 06:56 by tungld

Extend the verifyInputTensors to shape information provided with compilation option

#3346 opened 2025-12-09 09:48 by chentong319

Profiling into a buffer

#3342 opened 2025-12-03 19:09 by AlexandreEichenberger

Add CodeQL analysis workflow configuration

#3335 opened 2025-11-13 20:59 by Sunny-Anand

Do not generate stablehlo.dynamic_broadcast_in_dim when shapes are same.

#3328 opened 2025-11-04 08:09 by nikhil-m-k

Fix multi-config generator handling and detection

#3315 opened 2025-10-17 14:02 by jonasehrlich

Apply 'EXCLUDE_FROM_ALL' to the directory when ONNX_MLIR_BUILD_TESTS is not ON

#3287 opened 2025-09-25 19:45 by mschofie

Add minimal ONNX dialect build option

#3263 opened 2025-08-21 19:41 by anuragsingh-tt

Better handling of constants in -profile-ir

#3257 opened 2025-08-13 19:47 by AlexandreEichenberger

Fixes for building onnx-mlir on Windows with Visual Studio 2022

#3253 opened 2025-08-07 01:02 by mschofie

Stick/unstick fp32 data alloc at 4k pages

#3250 opened 2025-07-31 13:12 by AlexandreEichenberger

onnx update

#3238 opened 2025-07-23 18:04 by Sunny-Anand

Modify onnx.Cast(to:f16) to onnx.Cast(to:f32)

#3229 opened 2025-07-16 14:04 by chentong319

[Draft] Replace usage of applyPartialConversion with applyPatternsGreedily in Decompose.cpp

#3212 opened 2025-07-01 15:15 by jorickert

Propagate onnx.dim_params in shape inference

#3160 opened 2025-05-13 13:17 by chentong319

Backward fold scale axis to Gemm layer in ONNX Dialect

#3131 opened 2025-04-18 09:38 by Arkar-Hema

Converting Split->conv->concat to Grouped conv

#3124 opened 2025-04-16 11:18 by kumarappan-cmyk

Combine parallel dense Optimization pass in ONNX Dialect

#3123 opened 2025-04-16 08:34 by Arkar-Hema

Update PyExecutionSessionBase.cpp to get right result

#3040 opened 2025-01-10 05:47 by fairytale0828

Disabled building tests default (like LLVM), enabled for our CIs

#3021 opened 2024-12-02 22:01 by AlexandreEichenberger

Release MLIRContext and Module before executing opt command for memory reduction

#3000 opened 2024-11-05 08:07 by imaihal

Definition and shape inference for ONNXParallelOp and ONNXForkOp

#2810 opened 2024-05-01 04:20 by imaihal

[WIP] High-Level (Coase grain) parallelization with ONNXParallelOp and ONNXForkOp based on OpenMP

#2756 opened 2024-03-15 08:42 by imaihal

Loop collapsing with fusion

#2618 opened 2023-11-13 22:14 by chentong319

patch for --EmitJNI on windows

#2582 opened 2023-10-25 02:32 by armgong

[Experiment] remove flag to disable hybrid pass

#2576 opened 2023-10-24 01:40 by sorenlassen

[NNPA] Check return values of zdnn calls. (follow-up of PR#2267)

#2338 opened 2023-06-27 14:14 by imaihal

use the same mmap for all tensors from the same external data file

#2331 opened 2023-06-21 17:28 by sorenlassen