v1.1.0

What's Changed

• Merge Dev by @arjunsuresh in #296
• Fix postfix deps by @arjunsuresh in #321
• Merge Dev by @arjunsuresh in #320
• Fixes for aocc by @arjunsuresh in #324
• Modularize dmidecode meminfo parsing by @arjunsuresh in #325
• Support aocc download and install by @arjunsuresh in #326
• Sync Dev by @arjunsuresh in #327
• Add tests for mlc-scripts installation by @anandhu-eng in #323
• Fix AOCC install by @arjunsuresh in #329
• Support Intel SDE Tool, improved individual script tests by @arjunsuresh in #333
• Update meta.yaml by @arjunsuresh in #334
• Added tests to the generic script template by @arjunsuresh in #335
• Support flang in llvm-install by @arjunsuresh in #336
• Add option to skip the certificate check for rclone downloads by @anandhu-eng in #331
• Fixed meta for get-aocc by @arjunsuresh in #337
• Sync Dev by @arjunsuresh in #330
• Verify SSL moved to Script Automation by @anandhu-eng in #338
• Add dry run + generalise rclone download by @anandhu-eng in #339
• Change logging levels based on verbose and silent by @anandhu-eng in #305
• link submit-mlperf-results with generate-mlperf-inference-submission by @anandhu-eng in #340
• Handle submission base directory properly by @anandhu-eng in #342
• Support relative path in docker run by @anandhu-eng in #343
• Added kill process script by @arjunsuresh in #347
• Support base variation inside dynamic variations by @arjunsuresh in #348
• Fix for llvm version handling by @arjunsuresh in #350
• Fix for llvm version handling by @arjunsuresh in #351
• Fixes for Nvidia gptj by @arjunsuresh in #352
• Handle situation when cache is not present by @anandhu-eng in #356
• Merge from GO by @arjunsuresh in #358
• Support google-dns for nvidia gpt docker by @arjunsuresh in #361
• Fixes nvidia gptj model generation by @arjunsuresh in #362
• Fix for dmidecode + support mounting filepaths in docker by @anandhu-eng in #344
• Replace boolean usage by @anandhu-eng in #357
• Sync Dev by @arjunsuresh in #341
• Replace KSM with 1Password in test-mlperf-inference-tvm-resnet50.yml by @nathanw-mlc in #359
• Sync main by @arjunsuresh in #363
• Fix invalid format output error in fetch-secret job. by @nathanw-mlc in #366
• Support OpenAI API by @arjunsuresh in #368
• Added run files for openai call by @arjunsuresh in #369
• Sync main with dev by @anandhu-eng in #370
• Skip authentication when service account credentials are provided by @anandhu-eng in #371
• Fix import for OpenOrca by @anandhu-eng in #375
• Fix dgl version for mlperf inference rgat by @arjunsuresh in #377
• Path string fix by @anandhu-eng in #376
• Fix command generation by @anandhu-eng in #379
• fix command generation by @anandhu-eng in #380
• Fix command generation - paths with space by @anandhu-eng in #381
• Fix for handling space by @anandhu-eng in #383
• fixes for path issues by @anandhu-eng in #384
• Path str fix by @anandhu-eng in #385
• fix for path issue by @anandhu-eng in #386
• Fix for space in path by @anandhu-eng in #387
• Fixes the output path when there is space - compiler linkage by @anandhu-eng in #388
• Fix space in path issue for dump freeze by @anandhu-eng in #389
• Run benchmark with forked inference repo by @anandhu-eng in #390
• Corrected git repo link by @anandhu-eng in #391
• Fix for space in path - get generic python lib by @anandhu-eng in #392
• Fixes for R50 cpp GH action by @arjunsuresh in #394
• Update test-amd-mlperf-inference-implementations.yml by @arjunsuresh in #395
• Replace print with MLC Logger by @anandhu-eng in #396
• Use num_threads=1 for retinanet by @arjunsuresh in #397
• Added experiment to script automation by @anandhu-eng in #398
• add --multi-thread-streams=0 for rclone version >= 1.60.0 by @anandhu-eng in #402
• Fixes for llvm-install-src by @arjunsuresh in #404
• Sync Dev by @arjunsuresh in #378
• Improvements for install-gcc-src by @arjunsuresh in #405
• Contest 2025 gemini call by @H9660 in #403
• Fixes for oneapi by @arjunsuresh in #408
• Merge from GO by @arjunsuresh in #409
• Support mlc experiment script by @arjunsuresh in #411
• Support mlc experiment entries by @arjunsuresh in #412
• Support state info for experiment run by @arjunsuresh in #413
• Support exp_tags for experiments by @arjunsuresh in #414
• Fix nvidia-dali version for python3.8, fixes #410 by @arjunsuresh in #415
• Force str for version in script module by @arjunsuresh in #416
• Code changes for integrating nvidia v5.0 by @anandhu-eng in #417
• PyCuda version fix by @anandhu-eng in #418
• Fix for pycuda in nvidia-impl by @arjunsuresh in #420
• Deprecated version in app-mlperf-inference-nvidia-scripts by @anandhu-eng in #421
• Fixes for pycuda and versions by @anandhu-eng in #422
• Removed pycuda version fix by @arjunsuresh in #423
• Fix onnx version by @anandhu-eng in #425
• Fixes to docker mounts/user by @arjunsuresh in #426
• Update customize.py by @arjunsuresh in #427
• Support docker_build_env, fixes #424 by @arjunsuresh in #428
• Fix env export for get-mlperf-inference-src by @arjunsuresh in #430
• Fix ucx LD_LIBRARY_PATH for app-mlperf-inference-nvidia by @arjunsuresh in #432
• Added set-cpu-freq script by @arjunsuresh in #435
• Added get-lib-jemalloc by @arjunsuresh in #437
• Support mlc test for scripts needing PAT in github actions by @arjunsuresh in #438
• Added JEMALLOC_PATH for lib,jemalloc by @arjunsuresh in #439
• Add automotive reference implementation workflow by @anandhu-eng in #407
• Fix ipmitool version detect by @arjunsuresh in #443
• Fix detect-sudo by @arjunsuresh in #444
• Changes for automotive 0.5 by @anandhu-eng in #442
• Fix for early stopping and command generation by @anandhu-eng in #445
• Updated tags for cognata dataset dependency by @anandhu-eng in #447
• Merge from GO by @arjunsuresh in #448
• Sync Dev by @arjunsuresh in #440
• Fix missing dep by @anandhu-eng in #449
• Constantstream fixes + migrate submission generation tests by @anandhu-eng in #450
• Changes for supporting workflow call by @anandhu-eng in #451
• Fixes #454 by @anandhu-eng in #457
• Fixed type: fos to os by @H9660 in #459
• Migrating changes from inference pr2162 + compliance run code changes-automotive by @anandhu-eng in #452
• Add branching strategy and PR checklist by @anandhu-eng in #460
• Add test for automotive submission generation by @anandhu-eng in #461
• Add script for downloading DeepSeek R1 MLPerf dataset + Fix MLC Tests by @anandhu-eng in #462
• Prevent dev branch being passed explicitly to get inference repo by @anandhu-eng in #463
• Auto GPU detection + Updated run args generation for 5.0 by @anandhu-eng in #433
• Changes for NVIDIA 5.0 by @anandhu-eng in #466
• Remove default version for Nvidia common code by @anandhu-eng in #467
• Prevent milli second conversion for ConstantStream by @anandhu-eng in #472
• Changes for running submission generation test by @anandhu-eng in #469
• Merge from GO by @arjunsuresh in #455
• Cleanup READMEs by @arjunsuresh in #474
• Improve docs gen by @arjunsuresh in #475
• Improve docs generation by @arjunsuresh in #476
• Merge from GO by @arjunsuresh in #477
• Fix a bug in adr usage in script module by @arjunsuresh in #478
• Fix for inheriting base variations - Nvidia by @anandhu-eng in #480
• Support more configurations for lib-jemalloc by @arjunsuresh in #481
• Fix for oneapi install by @arjunsuresh in #483
• Improve mlc doc; support variations and example commands by @arjunsuresh in #486
• Add llama3.1-8b reference implementation by @anandhu-eng in #473
• Update GitHub action for v5.1 - submission generation by @anandhu-eng in #491
• Update meta.yaml | Remove hardwired branch for mlperf-inference-src by @arjunsuresh in #493
• Fix for non interactive rclone-GDrive downloads by @anandhu-eng in #489
• Update meta.yaml | Fix deeplabv3 rclone config by @arjunsuresh in #494
• add more logs n running test with modified meta by @anandhu-eng in #495
• fix gh action not picking up the modified files by @anandhu-eng in #497
• Fix typo by @anandhu-eng in #498
• Fix rclone config for abtf-ssd-pytorch by @arjunsuresh in #501
• Update submission generation gh action to include open division by @anandhu-eng in #500
• Support download of model and dataset through r2 downloader by @anandhu-eng in #496
• Merge from GO by @arjunsuresh in #504
• Fix resnet50 cuda run not using gpu by @anandhu-eng in #502
• Fix experiment run by @arjunsuresh in #505
• Fixes for nvidia mlperf inference by @arjunsuresh in #506
• Sync dev by @arjunsuresh in #487
• Update meta.yaml | Added default submission checker version by @arjunsuresh in #507
• Changes for supporting lint action by @anandhu-eng in #509
• Support simultaneous activation of multiple dynamic variations by @arjunsuresh in #512
• Support base variations inside combination of variations, fixes #38, support list in update_tags_from_env by @arjunsuresh in #513
• Fix gcc installed path by @arjunsuresh in #518
• Fix aocc version check regular expression by @arjunsuresh in #521
• Return generated readme paths by @anandhu-eng in #515
• fix for issue 479 by @anandhu-eng in #516
• Update PR checklist by @anandhu-eng in #514
• Merge from GO by @arjunsuresh in #524
• Fix backup url env by @anandhu-eng in #526
• Update document-scripts.yml | Use venv for document script by @arjunsuresh in #525
• Install ca-certificates on launching container with MLC by @anandhu-eng in #531
• Migrate dataset and model downloads to R2 by @anandhu-eng in #532
• Fix docker root run, numactl detect by @arjunsuresh in #534
• Cleanup docs by @arjunsuresh in #535
• Update document-scripts.yml: don't commit unless there are changes to… by @arjunsuresh in #537
• Improve docs template by @arjunsuresh in #538
• Fix for whisper reference benchmark by @anandhu-eng in #540
• Add script to detect/install apptainer by @anandhu-eng in #536
• Output the submission directory in push to GitHub by @anandhu-eng in #533
• Modifications for llama3.1-8b edge by @anandhu-eng in #539
• Add initial draft of help.py by @arjunsuresh in #542
• Improve script help by @arjunsuresh in #543
• Add logging in push results to GitHub by @anandhu-eng in #541
• Fix for ignoring incorrect tmp cache entries by @arjunsuresh in #545
• Update mlperf inference docker base image by @arjunsuresh in #547
• Remove version fix for pycuda, support cuda 12.8, 12.9 by @arjunsuresh in #549
• Fixes for whisper, dont print python env at end by @arjunsuresh in #551
• Changes to support R2 Downloader by @anandhu-eng in #548
• Fix typo by @anandhu-eng in #552
• llama3.1 -> llama3_1 by @anandhu-eng in #554
• Fix Waymo dataset extraction by @anandhu-eng in #555
• Fixes for mlperf-inference-v5.1 by @arjunsuresh in #556
• Fix typo by @anandhu-eng in #557
• Check required files before generating submission by @anandhu-eng in #559
• Support docker runs for install-llvm-src,install-gcc-src by @arjunsuresh in #562
• update llama3.1 8b edge dataset link - rclone by @anandhu-eng in #561
• Update customize.py | Fix docker error on gh actions by @arjunsuresh in #563
• Fix permission error for fake runs by @arjunsuresh in #564
• Fix default input keys in docker run command by @arjunsuresh in #565
• Fix nvidia-mlperf-inference-v5.0-dev by @arjunsuresh in #567
• Added support for Whisper external model download (#544) by @kamieyy in #546
• Support softlinks in get,gcc path variation by @arjunsuresh in #568
• Adding support to intel oneapi 2025.2.0 by @hemanthtanguturi1 in #569
• Export src repo hash for install gcc,llvm by @arjunsuresh in #570
• Support target for gcc,llvm build from src by @amd-arsuresh in #573
• Added --host option for install-gcc-src by @arjunsuresh in #574
• Support env in os_info, cross-compilation for llvm by @arjunsuresh in #575
• Update load-secrets-action to v3 to directly support Windows runners by @nathanw-mlc in #576
• Merge from GO: fixes for --help by @arjunsuresh in #577
• Removed protobuf version requirement for mlperf-inference resnet50 by @arjunsuresh in #579
• Support install,gcc on macos, fortran support for oneapi2025.2 by @arjunsuresh in #580
• Create codeql.yml by @arjunsuresh in #581
• Sync Dev by @arjunsuresh in #582
• Delete .github/workflows/test-mlperf-inference-rnnt.yml by @arjunsuresh in #584
• Support llvm-install-from-src on Macbook, make gh actions code safe by @arjunsuresh in #587
• Potential fix for code scanning alert no. 13: Code injection by @arjunsuresh in #585
• Potential fix for code scanning alert no. 12: Code injection by @arjunsuresh in #586
• Removed unused nvidia-generate-engine script by @arjunsuresh in #589
• Merge dev by @arjunsuresh in #588
• Merge from go by @arjunsuresh in #590
• Merge Dev, fixes for CodeQL scan by @arjunsuresh in #591
• Potential fix for code scanning alert no. 3: Code injection by @arjunsuresh in #593
• Replace wmic for Windows by @arjunsuresh in #607
• Changes for supporting Llama2-70b - r5.0 by @anandhu-eng in #604
• fix indentation by @anandhu-eng in #610
• Support default_env, gh_token for docker, relative path bug for --outdirname by @arjunsuresh in #613
• Support uploaded quantised models by @anandhu-eng in #612
• Support flang for aocc compiler by @arjunsuresh in #614
• Support cache expiration by @arjunsuresh in #617
• support quantisation with v5.1 hashes + env variable fixes by @anandhu-eng in #619
• Support cache with expiration for detect-os and detect-cpu by @arjunsuresh in #618
• Make --docker internal, and mlc docker script the default usage for docker runs by @arjunsuresh in #620
• Increment version for pypi package update by @anandhu-eng in #621
• Sync Dev by @anandhu-eng in #615

New Contributors

• @nathanw-mlc made their first contribution in #359
• @H9660 made their first contribution in #403
• @kamieyy made their first contribution in #546
• @hemanthtanguturi1 made their first contribution in #569
• @amd-arsuresh made their first contribution in #573

Full Changelog: v1.0.1...v1.1.0