Update download instructions to use MLCommons R2 downloader with correct URIs

- Remove rclone-based download instructions
- Replace .json URLs with correct .uri files from metadata directory
- Update download commands for DeepSeek-R1, Llama 3.1 8b, and Whisper
- Use new MLCommons downloader infrastructure
- Remove file size information from download instructions

Author: anivar

language/deepseek-r1/README.md

@@ -37,16 +37,16 @@ Download the preprocessed dataset using the MLCommons downloader:
 
 ```bash
 bash <(curl -s https://raw.githubusercontent.com/mlcommons/r2-downloader/refs/heads/main/mlc-r2-downloader.sh) \
-  https://inference.mlcommons-storage.org/deepseek-r1%2Fdataset.json
+  https://inference.mlcommons-storage.org/metadata/deepseek-r1-datasets-fp8-eval.uri
 ```
 
-This will download the dataset file `mlperf_deepseek_r1_dataset_4388_fp8_eval.pkl` (~163MB).
+This will download the dataset file `mlperf_deepseek_r1_dataset_4388_fp8_eval.pkl`.
 
 To specify a custom download directory, use the `-d` flag:
 ```bash
 bash <(curl -s https://raw.githubusercontent.com/mlcommons/r2-downloader/refs/heads/main/mlc-r2-downloader.sh) \
   -d /path/to/download/directory \
-  https://inference.mlcommons-storage.org/deepseek-r1%2Fdataset.json
+  https://inference.mlcommons-storage.org/metadata/deepseek-r1-datasets-fp8-eval.uri
 ```
 
 ### Calibration
@@ -63,7 +63,7 @@ Download the calibration dataset using the MLCommons downloader:
 
 ```bash
 bash <(curl -s https://raw.githubusercontent.com/mlcommons/r2-downloader/refs/heads/main/mlc-r2-downloader.sh) \
-  https://inference.mlcommons-storage.org/deepseek-r1%2Fcalibration.json
+  https://inference.mlcommons-storage.org/metadata/deepseek-r1-0528.uri
 ```
 
 This will download the calibration dataset file `mlperf_deepseek_r1_calibration_dataset_500_fp8_eval.pkl`.
@@ -72,7 +72,7 @@ To specify a custom download directory, use the `-d` flag:
 ```bash
 bash <(curl -s https://raw.githubusercontent.com/mlcommons/r2-downloader/refs/heads/main/mlc-r2-downloader.sh) \
   -d /path/to/download/directory \
-  https://inference.mlcommons-storage.org/deepseek-r1%2Fcalibration.json
+  https://inference.mlcommons-storage.org/metadata/deepseek-r1-0528.uri
 ```
 
 ## Docker

language/llama3.1-8b/README.md

@@ -149,9 +149,9 @@ mlcr get,dataset,cnndm,_validation,_datacenter,_llama3,_mlc,_rclone --outdirname
 **Native method**
 ```bash
 bash <(curl -s https://raw.githubusercontent.com/mlcommons/r2-downloader/refs/heads/main/mlc-r2-downloader.sh) \
-  https://inference.mlcommons-storage.org/llama3.1-8b%2Fcnn-eval.json
+  https://inference.mlcommons-storage.org/metadata/llama3-1-8b-cnn-eval.uri
 ```
-This will download `cnn_eval.json` (~267MB).
+This will download `cnn_eval.json`.
 
 #### 5000 samples (edge)
 
@@ -163,9 +163,9 @@ mlcr get,dataset,cnndm,_validation,_edge,_llama3,_mlc,_rclone --outdirname=<path
 **Native method**
 ```bash
 bash <(curl -s https://raw.githubusercontent.com/mlcommons/r2-downloader/refs/heads/main/mlc-r2-downloader.sh) \
-  https://inference.mlcommons-storage.org/llama3.1-8b%2Fsample-cnn-eval-5000.json
+  https://inference.mlcommons-storage.org/metadata/llama3-1-8b-sample-cnn-eval-5000.uri
 ```
-This will download `sample_cnn_eval_5000.json` (~95MB).
+This will download `sample_cnn_eval_5000.json`.
 
 #### Calibration
 
@@ -177,9 +177,9 @@ mlcr get,dataset,cnndm,_calibration,_llama3,_mlc,_rclone --outdirname=<path to d
 **Native method**
 ```bash
 bash <(curl -s https://raw.githubusercontent.com/mlcommons/r2-downloader/refs/heads/main/mlc-r2-downloader.sh) \
-  https://inference.mlcommons-storage.org/llama3.1-8b%2Fcnn-dailymail-calibration.json
+  https://inference.mlcommons-storage.org/metadata/llama3-1-8b-cnn-dailymail-calibration.uri
 ```
-This will download `cnn_dailymail_calibration.json` (~21MB).
+This will download `cnn_dailymail_calibration.json`.
 
 To specify a custom download directory for any of these, use the `-d` flag:
 ```bash

speech2text/README.md

@@ -111,16 +111,16 @@ Download the Whisper model using the MLCommons downloader:
 
 ```bash
 bash <(curl -s https://raw.githubusercontent.com/mlcommons/r2-downloader/refs/heads/main/mlc-r2-downloader.sh) \
-  https://inference.mlcommons-storage.org/whisper%2Fmodel.json
+  https://inference.mlcommons-storage.org/metadata/whisper-model.uri
 ```
 
-This will download the Whisper model files (~25GB).
+This will download the Whisper model files.
 
 To specify a custom download directory, use the `-d` flag:
 ```bash
 bash <(curl -s https://raw.githubusercontent.com/mlcommons/r2-downloader/refs/heads/main/mlc-r2-downloader.sh) \
   -d /path/to/download/directory \
-  https://inference.mlcommons-storage.org/whisper%2Fmodel.json
+  https://inference.mlcommons-storage.org/metadata/whisper-model.uri
 ```
 
 ### External Download (Not recommended for official submission)
@@ -161,16 +161,16 @@ Download the preprocessed dataset using the MLCommons downloader:
 
 ```bash
 bash <(curl -s https://raw.githubusercontent.com/mlcommons/r2-downloader/refs/heads/main/mlc-r2-downloader.sh) \
-  https://inference.mlcommons-storage.org/whisper%2Fdataset.json
+  https://inference.mlcommons-storage.org/metadata/whisper-dataset.uri
 ```
 
-This will download the LibriSpeech dataset files (~4.6GB).
+This will download the LibriSpeech dataset files.
 
 To specify a custom download directory, use the `-d` flag:
 ```bash
 bash <(curl -s https://raw.githubusercontent.com/mlcommons/r2-downloader/refs/heads/main/mlc-r2-downloader.sh) \
   -d /path/to/download/directory \
-  https://inference.mlcommons-storage.org/whisper%2Fdataset.json
+  https://inference.mlcommons-storage.org/metadata/whisper-dataset.uri
 ```
 
 ### Unprocessed