AWS sagemaker documentation change
Summary
Removed detailed technical content about Nova model recipes and replaced with redirect to updated documentation locations
Security assessment
This change relocates documentation without introducing or modifying security-related content. The removal of previous permissions examples (S3:GetObject on model-customization-recipes) appears to be part of general content reorganization rather than addressing a specific security vulnerability
Diff
diff --git a/sagemaker/latest/dg/nova-model-recipes.md b/sagemaker/latest/dg/nova-model-recipes.md index 9d3f142a6..bdee5c648 100644 --- a//sagemaker/latest/dg/nova-model-recipes.md +++ b//sagemaker/latest/dg/nova-model-recipes.md @@ -5,2 +4,0 @@ -Getting base recipesAvailable models and algorithmsAmazon Nova recipes reference - @@ -9,165 +7 @@ Getting base recipesAvailable models and algorithmsAmazon Nova recipes reference -You can get an Amazon Nova recipe from the [SageMaker HyperPod recipes](https://github.com/aws/sagemaker-hyperpod-recipes/tree/main/recipes_collection/recipes) repository. A Nova recipe is a [YAML](https://yaml.org/) configuration file that provides details to SageMaker AI on how to run your model customization job. It provides the base model name, sets training hyperparameters, defines optimization settings, and includes any additional options required to fine-tune or train the model successfully. - -You can also access Nova recipes through Amazon SageMaker Studio and Amazon SageMaker Unified Studio by navigating to the JumpStart model hub, selecting AWS, and browsing Amazon Nova models to find their associated recipes. Both Amazon SageMaker Studio and Amazon SageMaker Unified Studio provide sample notebooks for each recipe, which include all the necessary steps to modify the recipes and run customization jobs using SageMaker AI training jobs or Amazon SageMaker HyperPod environments. - -To access the recipes pages in Amazon SageMaker Studio, the execution role must have the following permissions. - -JSON - - -**** - - - - { - "Version":"2012-10-17", - "Statement": [ - { - "Effect": "Allow", - "Action": [ - "s3:GetObject" - ], - "Resource": [ - "arn:aws:s3:::*model-customization-recipes*" - ] - } - ] - } - - -To execute the sample notebooks on SageMaker training jobs and SageMaker HyperPod, use one of the following SageMaker distribution image versions: `2.7.1+`, `2.8.0+`, `3.2.1+`, `3.3.0+`. This applies to both Amazon SageMaker Studio and Amazon SageMaker Unified Studio. - -###### Topics - - * Getting Amazon Nova recipes - - * Available models and algorithms - - * Amazon Nova recipes reference - - - - -## Getting Amazon Nova recipes - -To get a base Amazon Nova recipe, clone the [SageMaker HyperPod recipes](https://github.com/aws/sagemaker-hyperpod-recipes/tree/main/recipes_collection/recipes) repository by running the following command. - - - git clone https://github.com/aws/sagemaker-hyperpod-recipes.git - -The base recipes are available at [`recipes_collection/recipes/`](https://github.com/aws/sagemaker-hyperpod-recipes/tree/main/recipes_collection/recipes/). - - - cd recipes_collection/recipes/ - -The Amazon Nova customization recipes are in the following folders. - -Recipe type | Folder ----|--- -SFT (Full-rank and PEFT), PPO, DPO (Full-rank and PEFT) | [fine-tuning/nova](https://github.com/aws/sagemaker-hyperpod-recipes/tree/main/recipes_collection/recipes/fine-tuning/nova) -Evaluation | [evaluation/nova](https://github.com/aws/sagemaker-hyperpod-recipes/tree/main/recipes_collection/recipes/evaluation/nova) -CPT | [training/nova](https://github.com/aws/sagemaker-hyperpod-recipes/tree/main/recipes_collection/recipes/training) - -## Available models and algorithms - -The following table summarizes the availability of customization for Amazon Nova models and supported algorithms using SageMaker AI. - -Model name | Model ID | Fine-tuning | Notes ----|---|---|--- -Amazon Nova Micro | amazon.nova-micro-v1:0:128k | Yes | For SFT, DPO, and RFT, this model accepts text as input and produces text as output only. -Amazon Nova Lite | amazon.nova-lite-v1:0:300k | Yes | - - * SFT - Accepts either text and/or image or text and/or video as the input and produces text as output. A single job can't combine text, image, and video in the same run. - * DPO - Accepts text and image as the input and produces text as output. - * RFT - Accepts text and image as input for single-turn conversations and produces text as output. - - -Amazon Nova Pro | amazon.nova-pro-v1:0:300k | Yes | - - * SFT - Accepts either text and/or image or text and/or video as the input and produces text as output. A single job can't combine text, image, and video in the same run. - * DPO - Accepts text and image as the input and produces text as output. - * RFT - Accepts text and image as input for single-turn conversations and produces text as output. - - -Amazon Nova Lite 2.0 | amazon.nova-lite-v2:0:300k | Yes | - - * SFT - Accepts either text and/or image or text and/or video as the input and produces text as output. A single job can't combine text, image, and video in the same run. - * DPO - Accepts text and image as the input and produces text as output. - * RFT - Accepts text and image as input for single-turn conversations and produces text as output. Improved reward optimization capabilities. - - - -## Amazon Nova recipes reference - -The tables below list detailed information of the Amazon Nova recipes reference. - -The table below lists detailed information of the Amazon Nova Lite recipes reference. - -Model | Category/Sub-category | Technique | Recipe Name | Image URI (Training Jobs) | Image URI (HyperPod) | Compute Instance ----|---|---|---|---|---|--- -Nova Lite | Training/Fine-tuning | Supervised Fine-Tuning (LoRA) | `nova_lite_p5_gpu_lora_sft.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-TJ-SFT-V2-latest` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-HP-SFT-V2-latest` | `ml.p5.48xlarge` -Nova Lite | Training/Fine-tuning | Supervised Fine-Tuning (Full) | `nova_lite_p5_gpu_sft.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-TJ-SFT-V2-latest` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-HP-SFT-V2-latest` | `ml.p5.48xlarge` -Nova Lite | Training/Fine-tuning | Direct Preference Optimization (Full) | `nova_lite_p5_gpu_dpo.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-TJ-DPO-V2-latest` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-HP-DPO-V2-latest` | `ml.p5.48xlarge` -Nova Lite | Training/Fine-tuning | Direct Preference Optimization (LoRA) | `nova_lite_p5_gpu_lora_dpo.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-TJ-DPO-V2-latest` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-HP-DPO-V2-latest` | `ml.p5.48xlarge` -Nova Lite | Training/Reinforcement learning | Reinforcement Fine-Tuning (RFT) | `nova_lite_rft.yaml` | n/a | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SMHP-PPO-TRAIN-latest` | `ml.p5.48xlarge` -Nova Lite | Training/Continued Pre-Training | Continued Pre-Training (base model) | `nova_lite_gpu_p5x16_pretrain.yaml` | n/a | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:HP-CPT-latest` | `ml.p5.48xlarge` -Nova Lite | Evaluation/Evaluate | Standard text benchmarks | `nova_lite_p5_48xl_general_text_benchmark_eval.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-evaluation-repo:SM-TJ-Eval-latest` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-evaluation-repo:SM-HP-Eval-latest` | `ml.p5.48xlarge` -Nova Lite | Evaluation/Evaluate | Custom dataset evaluation | `nova_lite_p5_48xl_bring_your_own_dataset_eval.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-evaluation-repo:SM-TJ-Eval-latest` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-evaluation-repo:SM-HP-Eval-latest` | `ml.p5.48xlarge` -Nova Lite | Evaluation/Evaluate | Multi-modal benchmarks | `nova_lite_p5_48_general_multi_modal_benchmark_eval.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-evaluation-repo:SM-TJ-Eval-latest` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-evaluation-repo:SM-HP-Eval-latest` | `ml.p5.48xlarge` -Nova Lite | Evaluation/Evaluate | LLM as a Judge | `nova_lite_p5_48xl_llm_judge_eval.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-evaluation-repo:SM-TJ-Eval-latest` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-evaluation-repo:SM-HP-Eval-latest` | `ml.p5.48xlarge` -Nova Lite | Training/Fine-tuning | Supervised Fine-Tuning (LoRA) | `nova_lite_g5_g6_12x_gpu_lora_sft.yaml` `nova_lite_g5_g6_48x_gpu_lora_sft.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-TJ-SFT-V2-latest` | n/a | `ml.g5.12xlarge`, `ml.g6.12xlarge`, `ml.g5.48xlarge`, `ml.g6.48xlarge` -Nova Lite | Training/Fine-tuning | DPO (LoRA) | `nova_lite_g5_g6_48x_gpu_lora_dpo.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-TJ-DPO-V2-latest` | n/a | `ml.g5.48xlarge`, `ml.g6.48xlarge` - -The table below lists detailed information of the Nova 2.0 Lite recipes reference. - -Model | Category/Sub-category | Technique | Recipe Name | Image URI (Training Jobs) | Image URI (HyperPod) | Compute Instance ----|---|---|---|---|---|--- -Nova 2.0 Lite | Training | Continued Pre Training | `nova_lite_2_0_p5x8_gpu_pretrain.yaml` | N/A | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-HP-CPT-V2-latest` | `ml.p5.48xlarge`, `ml.p5en.48xlarge` -Nova 2.0 Lite | Parameter Efficient Fine Tuning | Fine Tuning | `nova_lite_2_0_p5_gpu_lora_sft.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-TJ-SFT-V2-latest` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-HP-SFT-V2-latest` | `ml.p5.48xlarge`, `ml.p5en.48xlarge` -Nova 2.0 Lite | Full-Rank Fine Tuning | Fine Tuning | `nova_lite_2_0_p5_gpu_sft.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-TJ-SFT-V2-latest` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-HP-SFT-V2-latest` | `ml.p5.48xlarge`, `ml.p5en.48xlarge` -Nova 2.0 Lite | Parameter Efficient RFT | Fine Tuning | `nova_lite_v2_smtj_p5_p5en_gpu_lora_rft.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-TJ-RFT-V2-latest` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-HP-RFT-TRAIN-V2-latest` | `ml.p5.48xlarge`, `ml.p5en.48xlarge` -Nova 2.0 Lite | Full Rank RFT | Fine Tuning | `nova_lite_v2_smtj_p5en_gpu_rft.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-TJ-RFT-V2-latest` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-HP-RFT-TRAIN-V2-latest` | `ml.p5.48xlarge`, `ml.p5en.48xlarge` -Nova 2.0 Lite | Bring Your Own Data | Evaluation | `nova_lite_2_0_p5_48xl_gpu_bring_your_own_dataset_eval.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-evaluation-repo:SM-TJ-Eval-V2-latest` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-evaluation-repo:SM-HP-Eval-V2-latest` | `ml.p5.48xlarge`, `ml.p5en.48xlarge` -Nova 2.0 Lite | General Text Benchmark | Evaluation | `nova_lite_2_0_p5_48xl_gpu_general_text_benchmark_eval.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-evaluation-repo:SM-TJ-Eval-V2-latest` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-evaluation-repo:SM-HP-Eval-V2-latest` | `ml.p5.48xlarge`, `ml.p5en.48xlarge` -Nova 2.0 Lite | RFT Evaluation | Evaluation | `nova_lite_2_0_p5_48xl_gpu_rft_eval.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-evaluation-repo:SM-TJ-Eval-V2-latest` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-evaluation-repo:SM-HP-Eval-V2-latest` | `ml.p5.48xlarge`, `ml.p5en.48xlarge` -Nova 2.0 Lite | LLM Rubric Based Judge | Evaluation | `nova_lite_2_0_p5_48xl_gpu_rubric_llm_judge_eval.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-evaluation-repo:SM-TJ-Eval-V2-latest` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-evaluation-repo:SM-HP-Eval-V2-latest` | `ml.p5.48xlarge`, `ml.p5en.48xlarge` - -The table below lists detailed information of the Amazon Nova Micro recipes reference. - -Model | Category/Sub-category | Technique | Recipe Name | Image URI (Training Jobs) | Image URI (HyperPod) | Compute Instance ----|---|---|---|---|---|--- -Nova Micro | Training/Fine-tuning | Supervised Fine-Tuning (LoRA) | `nova_micro_g5_gpu_lora_sft.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-TJ-SFT-V2-latest` | n/a | `ml.g5.12xlarge` -Nova Micro | Training/Fine-tuning | Supervised Fine-Tuning (LoRA) | `nova_micro_p5_gpu_lora_sft.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-TJ-SFT-V2-latest` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-HP-SFT-V2-latest` | `ml.p5.48xlarge` -Nova Micro | Training/Fine-tuning | Supervised Fine-Tuning (Full) | `nova_micro_p5_gpu_sft.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-TJ-SFT-V2-latest` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-HP-SFT-V2-latest` | `ml.p5.48xlarge` -Nova Micro | Training/Fine-tuning | Direct Preference Optimization (Full) | `nova_micro_p5_gpu_dpo.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-TJ-DPO-V2-latest` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-HP-DPO-V2-latest` | `ml.p5.48xlarge` -Nova Micro | Training/Fine-tuning | Direct Preference Optimization (LoRA) | `nova_micro_p5_gpu_lora_dpo.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-TJ-DPO-V2-latest` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-HP-DPO-V2-latest` | `ml.p5.48xlarge` -Nova Micro | Training/Reinforcement learning | Reinforcement Fine-Tuning (RFT) | `nova_micro_rft.yaml` | n/a | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SMHP-PPO-TRAIN-latest` | `ml.p5.48xlarge` -Nova Micro | Training/Continued Pre-Training | Continued Pre-Training (Base Model) | `nova_micro_gpu_p5x8_pretrain.yaml` | n/a | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:HP-CPT-latest` | `ml.p5.48xlarge` -Nova Micro | Evaluation/Evaluate | General text benchmark | `nova_micro_p5_48xl_general_text_benchmark_eval.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-evaluation-repo:SM-TJ-Eval-latest` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-evaluation-repo:SM-HP-Eval-latest` | `ml.p5.48xlarge` -Nova Micro | Evaluation/Evaluate | Bring your own dataset (gen_qa) benchmark | `nova_micro_p5_48xl_bring_your_own_dataset_eval.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-evaluation-repo:SM-TJ-Eval-latest` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-evaluation-repo:SM-HP-Eval-latest` | `ml.p5.48xlarge` -Nova Micro | Evaluation/Evaluate | LLM as a Judge | `nova_micro_p5_48xl_llm_judge_eval.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-evaluation-repo:SM-TJ-Eval-latest` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-evaluation-repo:SM-HP-Eval-latest` | `ml.p5.48xlarge` -Nova Micro | Training/Fine-tuning | Supervised Fine-Tuning (LoRA) | `nova_micro_g5_g6_12x_gpu_lora_sft.yaml` `nova_micro_g5_g6_48x_gpu_lora_sft.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-TJ-SFT-V2-latest` | n/a | `ml.g5.12xlarge`, `ml.g6.12xlarge`, `ml.g5.48xlarge`, `ml.g6.48xlarge` -Nova Micro | Training/Fine-tuning | Supervised Fine-Tuning (FFT) | `nova_micro_g5_g6_48x_gpu_sft.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-TJ-SFT-V2-latest` | n/a | `ml.g5.48xlarge`, `ml.g6.48xlarge` -Nova Micro | Training/Fine-tuning | DPO (LoRA) | `nova_micro_g5_g6_12x_gpu_lora_dpo.yaml` `nova_micro_g5_g6_48x_gpu_lora_dpo.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-TJ-DPO-V2-latest` | n/a | `ml.g5.12xlarge`, `ml.g6.12xlarge`, `ml.g5.48xlarge`, `ml.g6.48xlarge` - -The table below lists detailed information of the Amazon Nova Pro recipes reference. - -Model | Category/Sub-category | Technique | Recipe Name | Image URI (Training Jobs) | Image URI (HyperPod) | Compute Instance ----|---|---|---|---|---|--- -Nova Pro | Training/Fine-tuning | Supervised Fine-Tuning (LoRA) | `nova_pro_p5_gpu_lora_sft.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-TJ-SFT-V2-latest` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-HP-SFT-V2-latest` | `ml.p5.48xlarge` -Nova Pro | Training/Fine-tuning | Supervised Fine-Tuning (Full) | `nova_pro_p5_gpu_sft.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-TJ-SFT-V2-latest` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-HP-SFT-V2-latest` | `ml.p5.48xlarge` -Nova Pro | Training/Fine-tuning | Direct Preference Optimization (Full) | `nova_pro_p5_gpu_dpo.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-TJ-DPO-V2-latest` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-HP-DPO-V2-latest` | `ml.p5.48xlarge` -Nova Pro | Training/Fine-tuning | Direct Preference Optimization (LoRA) | `nova_pro_p5_gpu_lora_dpo.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-TJ-DPO-V2-latest` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SM-HP-DPO-V2-latest` | `ml.p5.48xlarge` -Nova Pro | Training/Reinforcement learning | Reinforcement Fine-Tuning (RFT) | `nova_pro_rft.yaml` | n/a | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:SMHP-PPO-TRAIN-latest` | `ml.p5.48xlarge` -Nova Pro | Training/Continued Pre-Training | Continued Pre-Training (Base Model) | `nova_pro_gpu_p5x24_pretrain.yaml` | n/a | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-fine-tune-repo:HP-CPT-latest` | `ml.p5.48xlarge` -Nova Pro | Training/Data augmentation | Model distillation for post-training | `nova_pro_r5_cpu_distill.yaml` | n/a | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-distillation-repo:SM-TJ-DISTILL-LATEST` | `ml.r5.24xlarge` -Nova Pro | Evaluation/Evaluate | Standard text benchmarks | `nova_pro_p5_48xl_general_text_benchmark_eval.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-evaluation-repo:SM-TJ-Eval-latest` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-evaluation-repo:SM-HP-Eval-latest` | `ml.p5.48xlarge` -Nova Pro | Evaluation/Evaluate | Custom dataset evaluation | `nova_pro_p5_48xl_bring_your_own_dataset_eval.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-evaluation-repo:SM-TJ-Eval-latest` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-evaluation-repo:SM-HP-Eval-latest` | `ml.p5.48xlarge` -Nova Pro | Evaluation/Evaluate | Multi-modal benchmarks | `nova_pro_p5_48xl_general_multi_modal_benchmark_eval.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-evaluation-repo:SM-TJ-Eval-latest` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-evaluation-repo:SM-HP-Eval-latest` | `ml.p5.48xlarge` -Nova Pro | Evaluation/Evaluate | LLM as a Judge | `nova_pro_p5_48xl_llm_judge_eval.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-evaluation-repo:SM-TJ-Eval-latest` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-evaluation-repo:SM-HP-Eval-latest` | `ml.p5.48xlarge` - -The table below lists detailed information of the Amazon Nova Premier recipes reference. - -Model | Category/Sub-category | Technique | Recipe Name | Image URI (Training Jobs) | Image URI (SageMaker HyperPod) | Compute Instance ----|---|---|---|---|---|--- -Nova Premier | Training | Model distillation for post-training | `nova_premier_r5_cpu_distill.yaml` | `708977205387.dkr.ecr.us-east-1.amazonaws.com/nova-distillation-repo:SM-TJ-DISTILL-LATEST` | n/a | `ml.r5.24xlarge` +This topic has moved. For the latest information, see Amazon Nova recipes in the [Amazon Nova 1.0 user guide](https://docs.aws.amazon.com//nova/latest/userguide/nova-model-recipes.html) or the [Amazon Nova 2.0 user guide](https://docs.aws.amazon.com//nova/latest/nova2-userguide/nova-model-recipes.html). @@ -181 +15 @@ To use the Amazon Web Services Documentation, Javascript must be enabled. Please -General prerequisites +Amazon Nova model customization