From 031e1b715d9e2cc1f421dfc981ecf9806502c061 Mon Sep 17 00:00:00 2001 From: Nathan Wasson Date: Tue, 7 Oct 2025 12:23:03 -0500 Subject: [PATCH 01/15] Remove Rclone instructions from README.md --- graph/R-GAT/README.md | 21 +-------------------- 1 file changed, 1 insertion(+), 20 deletions(-) diff --git a/graph/R-GAT/README.md b/graph/R-GAT/README.md index 7e37b8eeaa..2545bfa894 100644 --- a/graph/R-GAT/README.md +++ b/graph/R-GAT/README.md @@ -85,31 +85,12 @@ mlcr get,ml-model,rgat,_r2-downloader,_mlcommons --outdirname= ### Download the model using MLC R2 Downloader -Download the model using the MLCommons R2 Downloader: +Download the model using the MLCommons R2 Downloader (More information about the MLC R2 Downloader, including how to run it on Windows, can be found [here](https://inference.mlcommons-storage.org).): ```bash bash <(curl -s https://raw.githubusercontent.com/mlcommons/r2-downloader/refs/heads/main/mlc-r2-downloader.sh) https://inference.mlcommons-storage.org/metadata/rgat-model.uri ``` -### Download model using Rclone - -To run Rclone on Windows, you can download the executable [here](https://rclone.org/install/#windows). -To install Rclone on Linux/macOS/BSD systems, run: -``` -sudo -v ; curl https://rclone.org/install.sh | sudo bash -``` -Once Rclone is installed, run the following command to authenticate with the bucket: -``` -rclone config create mlc-inference s3 provider=Cloudflare access_key_id=f65ba5eef400db161ea49967de89f47b secret_access_key=fbea333914c292b854f14d3fe232bad6c5407bf0ab1bebf78833c2b359bdfd2b endpoint=https://c2686074cb2caf5cbaf6d134bdba8b47.r2.cloudflarestorage.com -``` -You can then navigate in the terminal to your desired download directory and run the following commands to download the checkpoints: - -**`fp32`** -``` -rclone copy mlc-inference:mlcommons-inference-wg-public/R-GAT/RGAT.pt $MODEL_PATH -P -``` - - ### Download and setup dataset #### Debug Dataset From c4745a8de8d40e91d5f672da6f9193f42c9e3e84 Mon Sep 17 00:00:00 2001 From: Nathan Wasson Date: Tue, 7 Oct 2025 12:28:37 -0500 Subject: [PATCH 02/15] Remove Rclone download instructions from README.md --- language/llama3.1-405b/README.md | 28 ++-------------------------- 1 file changed, 2 insertions(+), 26 deletions(-) diff --git a/language/llama3.1-405b/README.md b/language/llama3.1-405b/README.md index 52d986628f..8d2e8bb0e4 100644 --- a/language/llama3.1-405b/README.md +++ b/language/llama3.1-405b/README.md @@ -101,7 +101,7 @@ pip install -e ../../loadgen ### MLCommons Members Download (Recommended for official submission) -MLCommons hosts the model for download **exclusively by MLCommons Members**. You must first agree to the [confidentiality notice](https://llama3-1.mlcommons.org) using your organizational email address, then you will receive a link to a directory containing Rclone download instructions. _If you cannot access the form but you are part of a MLCommons Member organization, submit the [MLCommons subscription form](https://mlcommons.org/community/subscribe/) with your organizational email address and [associate a Google account](https://accounts.google.com/SignUpWithoutGmail) with your organizational email address._ +MLCommons hosts the model for download **exclusively by MLCommons Members**. You must first agree to the [confidentiality notice](https://llama3-1.mlcommons.org) using your organizational email address, then you will receive a link to a page with download instructions. _If you cannot access the form but you are part of a MLCommons Member organization, submit the [MLCommons subscription form](https://mlcommons.org/community/subscribe/) with your organizational email address and [associate a Google account](https://accounts.google.com/SignUpWithoutGmail) with your organizational email address._ ### Download model through MLCFlow Automation @@ -152,7 +152,7 @@ mlcr get,dataset,mlperf,inference,llama3,_calibration,_r2-downloader --outdirnam **Using R2-Downloader** -Download the model using the MLCommons R2 Downloader: +Download the model using the MLCommons R2 Downloader (more information about the MLC R2 Downloader, including how to run it on Windows, can be found [here](https://inference.mlcommons-storage.org)): Validation: @@ -166,30 +166,6 @@ Calibration: bash <(curl -s https://raw.githubusercontent.com/mlcommons/r2-downloader/refs/heads/main/mlc-r2-downloader.sh) https://inference.mlcommons-storage.org/metadata/llama3-1-405b-calibration-dataset-512.uri ``` -**Using RClone** - -You can use Rclone to download the preprocessed dataset from a Cloudflare R2 bucket. - -To run Rclone on Windows, you can download the executable [here](https://rclone.org/install/#windows). -To install Rclone on Linux/macOS/BSD systems, run: -``` -sudo -v ; curl https://rclone.org/install.sh | sudo bash -``` -Once Rclone is installed, run the following command to authenticate with the bucket: -``` -rclone config create mlc-inference s3 provider=Cloudflare access_key_id=f65ba5eef400db161ea49967de89f47b secret_access_key=fbea333914c292b854f14d3fe232bad6c5407bf0ab1bebf78833c2b359bdfd2b endpoint=https://c2686074cb2caf5cbaf6d134bdba8b47.r2.cloudflarestorage.com -``` -You can then navigate in the terminal to your desired download directory and run the following command to download the dataset: - -``` -rclone copy mlc-inference:mlcommons-inference-wg-public/llama3.1_405b/mlperf_llama3.1_405b_dataset_8313_processed_fp16_eval.pkl ./ -P -``` - -You can also download the calibration dataset from the Cloudflare R2 bucket by running the following command: - -``` -rclone copy mlc-inference:mlcommons-inference-wg-public/llama3.1_405b/mlperf_llama3.1_405b_calibration_dataset_512_processed_fp16_eval.pkl ./ -P -``` ## Run Performance Benchmarks From 2936458c626f341c3a340caae0a9545fc7a89c86 Mon Sep 17 00:00:00 2001 From: Nathan Wasson Date: Tue, 7 Oct 2025 12:29:40 -0500 Subject: [PATCH 03/15] Tweak README.md --- graph/R-GAT/README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/graph/R-GAT/README.md b/graph/R-GAT/README.md index 2545bfa894..6a44989fca 100644 --- a/graph/R-GAT/README.md +++ b/graph/R-GAT/README.md @@ -85,7 +85,7 @@ mlcr get,ml-model,rgat,_r2-downloader,_mlcommons --outdirname= ### Download the model using MLC R2 Downloader -Download the model using the MLCommons R2 Downloader (More information about the MLC R2 Downloader, including how to run it on Windows, can be found [here](https://inference.mlcommons-storage.org).): +Download the model using the MLCommons R2 Downloader (more information about the MLC R2 Downloader, including how to run it on Windows, can be found [here](https://inference.mlcommons-storage.org)): ```bash bash <(curl -s https://raw.githubusercontent.com/mlcommons/r2-downloader/refs/heads/main/mlc-r2-downloader.sh) https://inference.mlcommons-storage.org/metadata/rgat-model.uri From 80042ce3e85a27a194163dcd825af513e11cc1a2 Mon Sep 17 00:00:00 2001 From: Nathan Wasson Date: Wed, 8 Oct 2025 15:30:17 -0500 Subject: [PATCH 04/15] Switch from Rclone to R2 Downloader in README.md --- recommendation/dlrm_v2/pytorch/README.md | 34 +++++++----------------- 1 file changed, 9 insertions(+), 25 deletions(-) diff --git a/recommendation/dlrm_v2/pytorch/README.md b/recommendation/dlrm_v2/pytorch/README.md index 9a0c523334..2d2e00e02b 100755 --- a/recommendation/dlrm_v2/pytorch/README.md +++ b/recommendation/dlrm_v2/pytorch/README.md @@ -81,27 +81,18 @@ CFLAGS="-std=c++14" python setup.py develop --user mlcr get,preprocessed,dataset,criteo,_validation --outdirname= -j ``` -#### Download the preprocessed dataset using Rclone. +#### Download the preprocessed dataset using the MLCommons R2 Downloader (more information about the MLC R2 Downloader, including how to run it on Windows, can be found [here](https://inference.mlcommons-storage.org)). -To run Rclone on Windows, you can download the executable [here](https://rclone.org/install/#windows). -To install Rclone on Linux/macOS/BSD systems, run: -``` -sudo -v ; curl https://rclone.org/install.sh | sudo bash -``` -Once Rclone is installed, run the following command to authenticate with the bucket: -``` -rclone config create mlc-inference s3 provider=Cloudflare access_key_id=f65ba5eef400db161ea49967de89f47b secret_access_key=fbea333914c292b854f14d3fe232bad6c5407bf0ab1bebf78833c2b359bdfd2b endpoint=https://c2686074cb2caf5cbaf6d134bdba8b47.r2.cloudflarestorage.com -``` Prepare your dataset destination: -``` +``` shell cd $HOME/mlcommons/inference/recommendation/dlrm_v2/pytorch/ mkdir ./dataset && cd ./dataset mv ./ export DATA_DIR=./dataset ``` Download the dataset -``` -rclone copy mlc-inference:mlcommons-inference-wg-public/dlrm_preprocessed ./dataset -P +``` bash +bash <(curl -s https://raw.githubusercontent.com/mlcommons/r2-downloader/refs/heads/main/mlc-r2-downloader.sh) -d ./dataset https://inference.mlcommons-storage.org/metadata/dlrm-v2-preprocessed-dataset.uri ``` @@ -120,21 +111,14 @@ mlcr get,ml-model,get,ml-model,dlrm,_pytorch,_weight_sharded,_rclone --outdirnam #### Manual method -The above command automatically runs a set of Rclone commands to download the data from a Cloudflare R2 bucket. However, if you'd like to run the Rclone commands manually, you can do so as follows: +The above command automatically runs a set of commands to download the data from a Cloudflare R2 bucket. However, if you'd like to run the commands manually, you can do so as follows: + +(More information about the MLC R2 Downloader, including how to run it on Windows, can be found [here](https://inference.mlcommons-storage.org)) -To run Rclone on Windows, you can download the executable [here](https://rclone.org/install/#windows). -To install Rclone on Linux/macOS/BSD systems, run: -``` -sudo -v ; curl https://rclone.org/install.sh | sudo bash -``` -Once Rclone is installed, run the following command to authenticate with the bucket: -``` -rclone config create mlc-inference s3 provider=Cloudflare access_key_id=f65ba5eef400db161ea49967de89f47b secret_access_key=fbea333914c292b854f14d3fe232bad6c5407bf0ab1bebf78833c2b359bdfd2b endpoint=https://c2686074cb2caf5cbaf6d134bdba8b47.r2.cloudflarestorage.com -``` You can then navigate in the terminal to your desired download directory and run the following command to download the model weights: -``` -rclone copy mlc-inference:mlcommons-inference-wg-public/model_weights ./model_weights -P +``` bash +bash <(curl -s https://raw.githubusercontent.com/mlcommons/r2-downloader/refs/heads/main/mlc-r2-downloader.sh) https://inference.mlcommons-storage.org/metadata/dlrm-v2-model-weights.uri ``` #### (optional) From 73bed9a856108cb85877e0d38621dc3cd31e6cfe Mon Sep 17 00:00:00 2001 From: Nathan Wasson Date: Wed, 8 Oct 2025 15:39:18 -0500 Subject: [PATCH 05/15] Switch from Rclone to R2 Downloader in README.md --- language/mixtral-8x7b/README.md | 41 ++++++++++++--------------------- 1 file changed, 15 insertions(+), 26 deletions(-) diff --git a/language/mixtral-8x7b/README.md b/language/mixtral-8x7b/README.md index ed3f857ff7..3d663fe5ed 100644 --- a/language/mixtral-8x7b/README.md +++ b/language/mixtral-8x7b/README.md @@ -78,21 +78,14 @@ mlcr get,ml-model,mixtral --outdirname= -j ### Get Checkpoint -#### Using Rclone +#### Using the MLCommons R2 Downloader -To run Rclone on Windows, you can download the executable [here](https://rclone.org/install/#windows). -To install Rclone on Linux/macOS/BSD systems, run: -``` -sudo -v ; curl https://rclone.org/install.sh | sudo bash -``` -Once Rclone is installed, run the following command to authenticate with the bucket: -``` -rclone config create mlc-inference s3 provider=Cloudflare access_key_id=f65ba5eef400db161ea49967de89f47b secret_access_key=fbea333914c292b854f14d3fe232bad6c5407bf0ab1bebf78833c2b359bdfd2b endpoint=https://c2686074cb2caf5cbaf6d134bdba8b47.r2.cloudflarestorage.com -``` -You can then navigate in the terminal to your desired download directory and run the following command to download the model checkpoint: +(More information about the MLC R2 Downloader, including how to run it on Windows, can be found [here](https://inference.mlcommons-storage.org)) -``` -rclone copy mlc-inference:mlcommons-inference-wg-public/mixtral_8x7b/mixtral-8x7b-instruct-v0.1 ./mixtral-8x7b-instruct-v0.1 -P +Navigate in the terminal to your desired download directory and run the following command to download the model checkpoint: + +``` bash +bash <(curl -s https://raw.githubusercontent.com/mlcommons/r2-downloader/refs/heads/main/mlc-r2-downloader.sh) https://inference.mlcommons-storage.org/metadata/mixtral-8x7b-model-checkpoint.uri ``` ## Get Dataset @@ -115,17 +108,12 @@ mlcr get,dataset-mixtral,openorca-mbxp-gsm8k-combined,_calibration --outdirname= ### Preprocessed -#### Using Rclone -We make many of the MLPerf infernce models and datasets available using Rclone. In order to keep compatibility, you can use Rclone to get the preprocessed dataset: +#### Using the MLCommons R2 Downloader +We make many of the MLPerf infernce models and datasets available using the MLC R2 Downloader (more information about the MLC R2 Downloader, including how to run it on Windows, can be found [here](https://inference.mlcommons-storage.org)). In order to keep compatibility, you can use the MLC R2 Downloader to get the preprocessed dataset: -To run Rclone on Windows, you can download the executable [here](https://rclone.org/install/#windows). -To install Rclone on Linux/macOS/BSD systems, run: -```bash -sudo -v ; curl https://rclone.org/install.sh | sudo bash -``` -Once Rclone is installed, cd into the folder where you want to place the dataset and run: +`cd` into the folder where you want to place the dataset and run: ```bash -rclone copyurl https://inference.mlcommons-storage.org/mixtral_8x7b/09292024_mixtral_15k_mintoken2_v1.pkl ./ -a -P +bash <(curl -s https://raw.githubusercontent.com/mlcommons/r2-downloader/refs/heads/main/mlc-r2-downloader.sh) https://inference.mlcommons-storage.org/metadata/mixtral-8x7b-validation-dataset.uri ``` #### Using wget @@ -138,10 +126,11 @@ wget https://inference.mlcommons-storage.org/mixtral_8x7b/09292024_mixtral_15k_m ### Calibration dataset -#### Using Rclone -Rclone is installed, cd into the folder where you want to place the dataset and run: +#### Using the MLCommons R2 Downloader + +`cd` into the folder where you want to place the dataset and run: ```bash -rclone copyurl https://inference.mlcommons-storage.org/mixtral_8x7b%2F2024.06.06_mixtral_15k_calibration_v4.pkl ./ -a -P +bash <(curl -s https://raw.githubusercontent.com/mlcommons/r2-downloader/refs/heads/main/mlc-r2-downloader.sh) https://inference.mlcommons-storage.org/metadata/mixtral-8x7b-calibration-dataset.uri ``` #### Using wget @@ -307,4 +296,4 @@ For official submissions, 99% of each reference score is enforced. Additionally, ## Automated command for submission generation via MLCFlow -Please see the [new docs site](https://docs.mlcommons.org/inference/submission/) for an automated way to generate submission through MLCFlow. \ No newline at end of file +Please see the [new docs site](https://docs.mlcommons.org/inference/submission/) for an automated way to generate submission through MLCFlow. From aedc83d2b66f722f2ab6ed8b964a4c2e844153c9 Mon Sep 17 00:00:00 2001 From: Nathan Wasson Date: Wed, 8 Oct 2025 15:47:36 -0500 Subject: [PATCH 06/15] Switch from Rclone to R2 Downloader in README.md --- language/llama2-70b/README.md | 17 ++++------------- 1 file changed, 4 insertions(+), 13 deletions(-) diff --git a/language/llama2-70b/README.md b/language/llama2-70b/README.md index 8bec240469..5b3335b39d 100644 --- a/language/llama2-70b/README.md +++ b/language/llama2-70b/README.md @@ -66,7 +66,7 @@ CPU-only setup, as well as any GPU versions for applicable libraries like PyTorc ## Get Model ### MLCommons Members Download -MLCommons hosts the model and preprocessed dataset for download **exclusively by MLCommons Members**. You must first agree to the [confidentiality notice](https://llama2.mlcommons.org) using your organizational email address, then you will receive a link to a directory containing Rclone download instructions. _If you cannot access the form but you are part of a MLCommons Member organization, submit the [MLCommons subscription form](https://mlcommons.org/community/subscribe/) with your organizational email address and [associate a Google account](https://accounts.google.com/SignUpWithoutGmail) with your organizational email address._ +MLCommons hosts the model and preprocessed dataset for download **exclusively by MLCommons Members**. You must first agree to the [confidentiality notice](https://llama2.mlcommons.org) using your organizational email address, then you will receive a link to a page containing download instructions. _If you cannot access the form but you are part of a MLCommons Member organization, submit the [MLCommons subscription form](https://mlcommons.org/community/subscribe/) with your organizational email address and [associate a Google account](https://accounts.google.com/SignUpWithoutGmail) with your organizational email address._ ### Download model through MLCFlow Automation @@ -117,21 +117,12 @@ mlcr get,dataset,openorca,_calibration --outdirname= -j ### Preprocessed -You can use Rclone to download the preprocessed dataset from a Cloudflare R2 bucket. +You can use the MLCommons R2 Downloader to download the preprocessed dataset from a Cloudflare R2 bucket (more information about the MLC R2 Downloader, including how to run it on Windows, can be found [here](https://inference.mlcommons-storage.org)). -To run Rclone on Windows, you can download the executable [here](https://rclone.org/install/#windows). -To install Rclone on Linux/macOS/BSD systems, run: -``` -sudo -v ; curl https://rclone.org/install.sh | sudo bash -``` -Once Rclone is installed, run the following command to authenticate with the bucket: -``` -rclone config create mlc-inference s3 provider=Cloudflare access_key_id=f65ba5eef400db161ea49967de89f47b secret_access_key=fbea333914c292b854f14d3fe232bad6c5407bf0ab1bebf78833c2b359bdfd2b endpoint=https://c2686074cb2caf5cbaf6d134bdba8b47.r2.cloudflarestorage.com -``` -You can then navigate in the terminal to your desired download directory and run the following command to download the dataset: +Navigate in the terminal to your desired download directory and run the following command to download the dataset: ``` -rclone copy mlc-inference:mlcommons-inference-wg-public/open_orca ./open_orca -P +bash <(curl -s https://raw.githubusercontent.com/mlcommons/r2-downloader/refs/heads/main/mlc-r2-downloader.sh) https://inference.mlcommons-storage.org/metadata/llama-2-70b-open-orca-dataset.uri ``` ### Unprocessed From d2c72ab81dd11fe87777cecd8b3ba23127645a99 Mon Sep 17 00:00:00 2001 From: Nathan Wasson Date: Wed, 8 Oct 2025 19:48:50 -0500 Subject: [PATCH 07/15] Switch Rclone for R2 Downloader in README.md --- language/gpt-j/README.md | 17 +++++------------ 1 file changed, 5 insertions(+), 12 deletions(-) diff --git a/language/gpt-j/README.md b/language/gpt-j/README.md index 9c952b65db..4dc55a8681 100644 --- a/language/gpt-j/README.md +++ b/language/gpt-j/README.md @@ -103,21 +103,14 @@ mlcr get,ml-model,gptj,_pytorch,_rclone ---outdirname =./model -P #### Manual method -The above command automatically runs a set of Rclone commands to download the data from a Cloudflare R2 bucket. However, if you'd like to run the Rclone commands manually, you can do so as follows: +The above command automatically runs a set of commands to download the data from a Cloudflare R2 bucket. However, if you'd like to run the commands manually, you can do so as follows: -To run Rclone on Windows, you can download the executable [here](https://rclone.org/install/#windows). -To install Rclone on Linux/macOS/BSD systems, run: -``` -sudo -v ; curl https://rclone.org/install.sh | sudo bash -``` -Once Rclone is installed, run the following command to authenticate with the bucket: -``` -rclone config create mlc-inference s3 provider=Cloudflare access_key_id=f65ba5eef400db161ea49967de89f47b secret_access_key=fbea333914c292b854f14d3fe232bad6c5407bf0ab1bebf78833c2b359bdfd2b endpoint=https://c2686074cb2caf5cbaf6d134bdba8b47.r2.cloudflarestorage.com -``` -You can then navigate in the terminal to your desired download directory and run the following command to download the model checkpoint: +(More information about the MLC R2 Downloader, including how to run it on Windows, can be found [here](https://inference.mlcommons-storage.org)) + +Navigate in the terminal to your desired download directory and run the following command to download the model checkpoint: ``` -rclone copy mlc-inference:mlcommons-inference-wg-public/gpt-j ./model -P +bash <(curl -s https://raw.githubusercontent.com/mlcommons/r2-downloader/refs/heads/main/mlc-r2-downloader.sh) -d model https://inference.mlcommons-storage.org/metadata/gpt-j-model-checkpoint.uri ``` From d610bf76218818f69dcb564908745b1acb829eac Mon Sep 17 00:00:00 2001 From: Nathan Wasson Date: Wed, 8 Oct 2025 19:54:31 -0500 Subject: [PATCH 08/15] Switch Rclone for R2 Downloader in README.md --- text_to_image/README.md | 19 ++++++------------- 1 file changed, 6 insertions(+), 13 deletions(-) diff --git a/text_to_image/README.md b/text_to_image/README.md index e4e4d75af5..4c723c49c3 100644 --- a/text_to_image/README.md +++ b/text_to_image/README.md @@ -62,29 +62,22 @@ mlcr get,ml-model,sdxl,_fp32,_rclone --outdirname=$MODEL_PATH ``` #### Manual method -The above command automatically runs a set of Rclone commands to download the data from a Cloudflare R2 bucket. However, if you'd like to run the Rclone commands manually, you can do so as follows: +The above command automatically runs a set of commands to download the data from a Cloudflare R2 bucket. However, if you'd like to run the commands manually, you can do so as follows: -To run Rclone on Windows, you can download the executable [here](https://rclone.org/install/#windows). -To install Rclone on Linux/macOS/BSD systems, run: -``` -sudo -v ; curl https://rclone.org/install.sh | sudo bash -``` -Once Rclone is installed, run the following command to authenticate with the bucket: -``` -rclone config create mlc-inference s3 provider=Cloudflare access_key_id=f65ba5eef400db161ea49967de89f47b secret_access_key=fbea333914c292b854f14d3fe232bad6c5407bf0ab1bebf78833c2b359bdfd2b endpoint=https://c2686074cb2caf5cbaf6d134bdba8b47.r2.cloudflarestorage.com -``` -You can then navigate in the terminal to your desired download directory and run the following commands to download the checkpoints: +(More information about the MLC R2 Downloader, including how to run it on Windows, can be found [here](https://inference.mlcommons-storage.org)) + +Navigate in the terminal to your desired download directory and run the following commands to download the checkpoints: ``` cd $MODEL_PATH ``` **`fp32`** ``` -rclone copy mlc-inference:mlcommons-inference-wg-public/stable_diffusion_fp32 $MODEL_PATH -P +bash <(curl -s https://raw.githubusercontent.com/mlcommons/r2-downloader/refs/heads/main/mlc-r2-downloader.sh) -d $MODEL_PATH https://inference.mlcommons-storage.org/metadata/stable-diffusion-xl-1-0-fp32-checkpoint.uri ``` **`fp16`** ``` -rclone copy mlc-inference:mlcommons-inference-wg-public/stable_diffusion_fp16 $MODEL_PATH -P +bash <(curl -s https://raw.githubusercontent.com/mlcommons/r2-downloader/refs/heads/main/mlc-r2-downloader.sh) -d $MODEL_PATH https://inference.mlcommons-storage.org/metadata/stable-diffusion-xl-1-0-fp16-checkpoint.uri ``` ### Download validation dataset From 0d0abc2e2b9d023fd0279141406004eac96adeb9 Mon Sep 17 00:00:00 2001 From: ANANDHU S <71482562+anandhu-eng@users.noreply.github.com> Date: Tue, 14 Oct 2025 22:27:44 +0530 Subject: [PATCH 09/15] Use r2 downloader for gpt j model download (#2365) --- language/gpt-j/README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/language/gpt-j/README.md b/language/gpt-j/README.md index 4dc55a8681..9538b3a686 100644 --- a/language/gpt-j/README.md +++ b/language/gpt-j/README.md @@ -9,7 +9,7 @@ You can also do `pip install mlc-scripts` and then use `mlcr` commands for downl ### Download model through MLCFlow Automation ``` -mlcr get,ml-model,gptj,_pytorch --outdirname= -j +mlcr get,ml-model,gptj,_pytorch,_fp32,_r2-downloader --outdirname= -j ``` ### Download dataset through MLCFlow Automation From 50537285f7a085bbda385183a36083b448ff1145 Mon Sep 17 00:00:00 2001 From: ANANDHU S <71482562+anandhu-eng@users.noreply.github.com> Date: Tue, 14 Oct 2025 22:28:48 +0530 Subject: [PATCH 10/15] Provide r2 download commands for mixtral model and datasets (#2364) --- language/mixtral-8x7b/README.md | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) diff --git a/language/mixtral-8x7b/README.md b/language/mixtral-8x7b/README.md index 3d663fe5ed..14d4c8a59b 100644 --- a/language/mixtral-8x7b/README.md +++ b/language/mixtral-8x7b/README.md @@ -73,7 +73,7 @@ CPU-only setup, as well as any GPU versions for applicable libraries like PyTorc ### Download model through MLCFlow Automation ``` -mlcr get,ml-model,mixtral --outdirname= -j +mlcr get,ml-model,mixtral,_r2-downloader,_mlc --outdirname= -j ``` ### Get Checkpoint @@ -95,17 +95,15 @@ bash <(curl -s https://raw.githubusercontent.com/mlcommons/r2-downloader/refs/he **Validation** ``` -mlcr get,dataset-mixtral,openorca-mbxp-gsm8k-combined,_validation --outdirname= -j +mlcr get,dataset-mixtral,openorca-mbxp-gsm8k-combined,_validation,_r2-downloader --outdirname= -j ``` **Calibration** ``` -mlcr get,dataset-mixtral,openorca-mbxp-gsm8k-combined,_calibration --outdirname= -j +mlcr get,dataset-mixtral,openorca-mbxp-gsm8k-combined,_calibration,_r2-downloader --outdirname= -j ``` -- Adding `_wget` tag to the run command will change the download tool from `rclone` to `wget`. - ### Preprocessed #### Using the MLCommons R2 Downloader From 63c9336bf0027a0bfc14306a51189e72fa868ee4 Mon Sep 17 00:00:00 2001 From: ANANDHU S <71482562+anandhu-eng@users.noreply.github.com> Date: Tue, 14 Oct 2025 22:30:04 +0530 Subject: [PATCH 11/15] Replace MLCFlow RClone command for criteo dataset with R2 (#2363) --- recommendation/dlrm_v2/pytorch/README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/recommendation/dlrm_v2/pytorch/README.md b/recommendation/dlrm_v2/pytorch/README.md index 2d2e00e02b..d56909afd3 100755 --- a/recommendation/dlrm_v2/pytorch/README.md +++ b/recommendation/dlrm_v2/pytorch/README.md @@ -78,7 +78,7 @@ CFLAGS="-std=c++14" python setup.py develop --user #### Download dataset through MLCFlow Automation ``` -mlcr get,preprocessed,dataset,criteo,_validation --outdirname= -j +mlcr get,preprocessed,dataset,criteo,_r2-downloader,_mlc,_validation --outdirname= -j ``` #### Download the preprocessed dataset using the MLCommons R2 Downloader (more information about the MLC R2 Downloader, including how to run it on Windows, can be found [here](https://inference.mlcommons-storage.org)). From 8bac8230a95c99cc090cd4e0c2416cfff5e9ffe5 Mon Sep 17 00:00:00 2001 From: ANANDHU S <71482562+anandhu-eng@users.noreply.github.com> Date: Tue, 14 Oct 2025 22:41:50 +0530 Subject: [PATCH 12/15] Deprecate MLCFlow rclone download command with r2 (#2362) --- recommendation/dlrm_v2/pytorch/README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/recommendation/dlrm_v2/pytorch/README.md b/recommendation/dlrm_v2/pytorch/README.md index d56909afd3..24a38a1ee0 100755 --- a/recommendation/dlrm_v2/pytorch/README.md +++ b/recommendation/dlrm_v2/pytorch/README.md @@ -106,7 +106,7 @@ N/A | pytorch | <2GB | - #### Download model through MLCFlow Automation ``` -mlcr get,ml-model,get,ml-model,dlrm,_pytorch,_weight_sharded,_rclone --outdirname= -j +mlcr get,ml-model,get,ml-model,dlrm,_pytorch,_fp32,_weight_sharded,_r2-downloader --outdirname= -j ``` #### Manual method From 1f5b2d17fe07cc17223b99affddb457c547e2e56 Mon Sep 17 00:00:00 2001 From: ANANDHU S <71482562+anandhu-eng@users.noreply.github.com> Date: Tue, 14 Oct 2025 22:42:10 +0530 Subject: [PATCH 13/15] Add instruction to download DeepSeek model through MLCflow (#2361) --- language/deepseek-r1/README.md | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/language/deepseek-r1/README.md b/language/deepseek-r1/README.md index a6c30a6155..3e9aa99a76 100644 --- a/language/deepseek-r1/README.md +++ b/language/deepseek-r1/README.md @@ -13,6 +13,14 @@ You can also do pip install mlc-scripts and then use `mlcr` commands for downloa - DeepSeek-R1 model is automatically downloaded as part of setup - Checkpoint conversion is done transparently when needed. +**Using MLCFlow Automation** + +Download the model using the MLCFlow Automation: + +``` +mlcr get,ml-model,deepseek-r1,_r2-downloader,_mlc --outdirname= -j +``` + **Using the MLC R2 Downloader** Download the model using the MLCommons R2 Downloader: From 1cfeadd3f7976183a0cbaa86d953905f457e81c0 Mon Sep 17 00:00:00 2001 From: "github-actions[bot]" Date: Tue, 14 Oct 2025 17:13:06 +0000 Subject: [PATCH 14/15] [Automated Commit] Format Codebase --- speech2text/accuracy_eval.py | 4 ++-- speech2text/reference_SUT.py | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/speech2text/accuracy_eval.py b/speech2text/accuracy_eval.py index eb6cc53299..c356ab6398 100644 --- a/speech2text/accuracy_eval.py +++ b/speech2text/accuracy_eval.py @@ -57,12 +57,12 @@ "x", "y", "z", - "'", + "'", "0", "1", "2", "3", - "4", + "4", "5", "6", "7", diff --git a/speech2text/reference_SUT.py b/speech2text/reference_SUT.py index 63d491a00f..0b2f02c490 100644 --- a/speech2text/reference_SUT.py +++ b/speech2text/reference_SUT.py @@ -90,12 +90,12 @@ def get_start_cores(start_cores="0"): "x", "y", "z", - "'", + "'", "0", "1", "2", "3", - "4", + "4", "5", "6", "7", From 167d6f15d80971e6166599b5cfcd725157366e74 Mon Sep 17 00:00:00 2001 From: nathanwasson <126016071+nathanwasson@users.noreply.github.com> Date: Tue, 14 Oct 2025 16:30:14 -0500 Subject: [PATCH 15/15] Trigger cla-check