This document is relevant for: Inf1, Inf2, Trn1, Trn1n

Previous Releases Notes (Neuron 2.x)

Neuron 2.17.0 (02/13/2024)

What’s New

Neuron 2.17 release improves small collective communication operators (smaller than 16MB) by up to 30%, which improves large language model (LLM) Inference performance by up to 10%. This release also includes improvements in Neuron Profiler and other minor enhancements and bug fixes.

For more detailed release notes of the new features and resolved issues, see Neuron Components Release Notes.

To learn about the model architectures currently supported on Inf1, Inf2, Trn1 and Trn1n instances, please see model_architecture_fit.

Neuron 2.16.1 (01/18/2024)

Patch release with compiler bug fixes, updates to Neuron Device Plugin and Neuron Kubernetes Scheduler .

Neuron 2.16.0 (12/21/2023)

What’s New

Neuron 2.16 adds support for Llama-2-70B training and inference, upgrades to PyTorch 2.1 (beta) and adds new support for PyTorch Lightning Trainer (beta) as well as performance improvements and adding Amazon Linux 2023 support.

Training highlights: NeuronX Distributed library LLM models training performance is improved by up to 15%. LLM model training user experience is improved by introducing support of PyTorch Lightning Trainer (beta), and a new model optimizer wrapper which will minimize the amount of changes needed to partition models using NeuronX Distributed primitives.

Inference highlights: PyTorch inference now allows to dynamically swap different fine-tuned weights for an already loaded model, as well as overall improvements of LLM inference throughput and latency with Transformers NeuronX. Two new reference model samples for LLama-2-70b and Mistral-7b model inference.

User experience: This release introduces two new capabilities: A new tool, Neuron Distributed Event Tracing (NDET) which improves debuggability, and the support of profiling collective communication operators in the Neuron Profiler tool.

More release content can be found in the table below and each component release notes.

What’s New	Details	Instances
Transformers NeuronX (transformers-neuronx) for Inference	[Beta] Support for Grouped Query Attention(GQA). See developer guide [Beta] Support for Llama-2-70b model inference using Grouped Query Attention. See tutorial [Beta] Support for Mistral-7B-Instruct-v0.1 model inference. See sample code See more at Transformers Neuron (transformers-neuronx) release notes	Inf2, Trn1/Trn1n
NeuronX Distributed (neuronx-distributed) for Training	[Beta] Support for PyTorch Lightning to train models using tensor parallelism and data parallelism . See api guide , developer guide and tutorial Support for Model and Optimizer Wrapper training API that handles the parallelization. See api guide and Developer guide for model and optimizer wrapper (neuronx-distributed ) New save_checkpoint and load_checkpoint APIs to save/load checkpoints during distributed training. See Developer guide for save/load checkpoint (neuronx-distributed ) Support for a new Query-Key-Value(QKV) module that provides the ability to replicate the Key Value heads and adds flexibility to use higher Tensor parallel degree during Training. See api guide and tutorial See more at Neuron Distributed Release Notes (neuronx-distributed)	Trn1/Trn1n
NeuronX Distributed (neuronx-distributed) for Inference	Support weight-deduplication amongst TP shards by giving ability to save weights separately than in NEFF files. See developer guide Llama-2-7B model inference script ([html] [notebook]) See more at Neuron Distributed Release Notes (neuronx-distributed) and API Reference Guide (neuronx-distributed )	Inf2,Trn1/Trn1n
PyTorch NeuronX (torch-neuronx)	[Beta]Support for] PyTorch 2.1. See Introducing PyTorch 2.1 Support . See llama-2-13b inference sample. Support to separate out model weights from NEFF files and new replace_weights API to replace the separated weights. See PyTorch Neuron (torch-neuronx) Weight Replacement API for Inference and PyTorch NeuronX Tracing API for Inference [Beta] Script for training stabilityai/stable-diffusion-2-1-base and runwayml/stable-diffusion-v1-5 models . See script [Beta] Script for training facebook/bart-large model. See script [Beta] Script for stabilityai/stable-diffusion-2-inpainting model inference. See script	Trn1/Trn1n,Inf2
Neuron Tools	New Neuron Distributed Event Tracing (NDET) tool to help visualize execution trace logs and diagnose errors in multi-node workloads. See Neuron Distributed Event Tracing (NDET) User Guide Support for multi-worker jobs in neuron-profile . See Neuron Profile User Guide See more at Neuron System Tools	Inf1/Inf2/Trn1/Trn1n
Documentation Updates	Added setup guide instructions for AL2023 OS. See Setup Guide Added announcement for name change of Neuron Components. See Announcing Name Change for Neuron Components Added announcement for End of Support for PyTorch 1.10 . See Announcing End of Support for PyTorch Neuron version 1.10 Added announcement for End of Support for PyTorch 2.0 Beta. See Announcing End of Support for PyTorch NeuronX version 2.0 (beta) See more at Neuron Documentation Release Notes	Inf1, Inf2, Trn1/Trn1n
Minor enhancements and bug fixes.	See Neuron Components Release Notes	Trn1/Trn1n , Inf2, Inf1
Known Issues and Limitations	See 2.16.0 Known Issues and Limitations	Trn1/Trn1n , Inf2, Inf1
Release Artifacts	see Release Artifacts	Trn1/Trn1n , Inf2, Inf1

2.16.0 Known Issues and Limitations

• We recommend running multi-node training jobs on AL2023 using Amazon EKS. Parallel Cluster currently does not support AL2023.

• There are known compiler issues impacting inference accuracy of certain model configurations of Llama-2-13b when amp = fp16 is used. If this issue is observed, amp=fp32 should be used as a work around. This issue will be addressed in future Neuron releases.

• Execution time reported in neuron-profile tool is sometimes in-accurate due to a bug in how the time is captured. The bug will be addressed in upcoming Neuron releases.

• See component release notes below for any additional known issues.

Neuron 2.15.2 (11/17/2023)

Patch release that fixes compiler issues related to performance when training using neuronx-nemo-megatron library.

Neuron 2.15.1 (11/09/2023)

Patch release to fix execution overhead issues in Neuron Runtime that were inadvertently introduced in 2.15 release.

Neuron 2.15.0 (10/26/2023)

What’s New

This release adds support for PyTorch 2.0 (Beta), increases performance for both training and inference workloads, adding ability to train models like Llama-2-70B using neuronx-distributed. With this release, we are also adding pipeline parallelism support for neuronx-distributed enabling full 3D parallelism support to easily scale training to large model sizes. Neuron 2.15 also introduces support for training resnet50, milesial/Pytorch-UNet and deepmind/vision-perceiver-conv models using torch-neuronx, as well as new sample code for flan-t5-xl model inference using neuronx-distributed, in addition to other performance optimizations, minor enhancements and bug fixes.

What’s New	Details	Instances
Neuron Distributed (neuronx-distributed) for Training	Pipeline parallelism support. See API Reference Guide (neuronx-distributed ) , Developer guide for Pipeline Parallelism (neuronx-distributed ) and Pipeline Parallelism Overview Llama-2-70B model training script (sample script) (tutorial) Mixed precision support. See Developer guide for Pipeline Parallelism (neuronx-distributed ) Support serialized checkpoint saving and loading using save_xser and load_xser parameters. See API Reference Guide (neuronx-distributed ) See more at Neuron Distributed Release Notes (neuronx-distributed)	Trn1/Trn1n
Neuron Distributed (neuronx-distributed) for Inference	flan-t5-xl model inference script (tutorial) See more at Neuron Distributed Release Notes (neuronx-distributed) and API Reference Guide (neuronx-distributed )	Inf2,Trn1/Trn1n
Transformers Neuron (transformers-neuronx) for Inference	Serialization support for Llama, Llama-2, GPT2 and BLOOM models . See developer guide and tutorial See more at Transformers Neuron (transformers-neuronx) release notes	Inf2, Trn1/Trn1n
PyTorch Neuron (torch-neuronx)	Introducing PyTorch 2.0 Beta support. See Introducing PyTorch 2.0 Support (End of Support) . See llama-2-7b training , bert training and t5-3b inference samples. Scripts for training resnet50[Beta] , milesial/Pytorch-UNet[Beta] and deepmind/vision-perceiver-conv[Beta] models.	Trn1/Trn1n,Inf2
AWS Neuron Reference for Nemo Megatron library (neuronx-nemo-megatron)	Llama-2-70B model training sample using pipeline parallelism and tensor parallelism ( tutorial ) GPT-NeoX-20B model training using pipeline parallelism and tensor parallelism See more at AWS Neuron Reference for Nemo Megatron(neuronx-nemo-megatron) Release Notes and neuronx-nemo-megatron github repo	Trn1/Trn1n
Neuron Compiler (neuronx-cc)	New llm-training option argument to --distribution_strategy compiler option for optimizations related to distributed training. See more at Neuron Compiler CLI Reference Guide (neuronx-cc) See more at Neuron Compiler (neuronx-cc) release notes	Inf2/Trn1/Trn1n
Neuron Tools	alltoall Collective Communication operation for intra node(with in the instance), previously released in Neuron Collectives v2.15.13, was added as a testable operation in nccom-test. See NCCOM-TEST User Guide See more at Neuron System Tools	Inf1/Inf2/Trn1/Trn1n
Documentation Updates	New App Note and Developer Guide about Activation memory reduction using sequence parallelism and activation recomputation in neuronx-distributed Added a new Model Samples and Tutorials summary page. See Model Samples and Tutorials Added Neuron SDK Classification guide. See Neuron Software Classification See more at Neuron Documentation Release Notes	Inf1, Inf2, Trn1/Trn1n
Minor enhancements and bug fixes.	See Neuron Components Release Notes	Trn1/Trn1n , Inf2, Inf1
Release Artifacts	see Release Artifacts	Trn1/Trn1n , Inf2, Inf1

For more detailed release notes of the new features and resolved issues, see Neuron Components Release Notes.

To learn about the model architectures currently supported on Inf1, Inf2, Trn1 and Trn1n instances, please see model_architecture_fit.

Neuron 2.14.1 (09/26/2023)

This is a patch release that fixes compiler issues in certain configurations of Llama and Llama-2 model inference using transformers-neuronx.

Note

There is still a known compiler issue for inference of some configurations of Llama and Llama-2 models that will be addressed in future Neuron release. Customers are advised to use --optlevel 1 (or -O1) compiler flag to mitigate this known compiler issue.

See Neuron Compiler CLI Reference Guide (neuronx-cc) on the usage of --optlevel 1 compiler flag. Please see more on the compiler fix and known issues in Neuron Compiler (neuronx-cc) release notes and Transformers Neuron (transformers-neuronx) release notes

Neuron 2.14.0 (09/15/2023)

What’s New

This release introduces support for Llama-2-7B model training and T5-3B model inference using neuronx-distributed. It also adds support for Llama-2-13B model training using neuronx-nemo-megatron. Neuron 2.14 also adds support for Stable Diffusion XL(Refiner and Base) model inference using torch-neuronx . This release also introduces other new features, performance optimizations, minor enhancements and bug fixes. This release introduces the following:

Note

This release deprecates --model-type=transformer-inference compiler flag. Users are highly encouraged to migrate to the --model-type=transformer compiler flag.

What’s New	Details	Instances
AWS Neuron Reference for Nemo Megatron library (neuronx-nemo-megatron)	Llama-2-13B model training support ( tutorial ) ZeRO-1 Optimizer support that works with tensor parallelism and pipeline parallelism See more at AWS Neuron Reference for Nemo Megatron(neuronx-nemo-megatron) Release Notes and neuronx-nemo-megatron github repo	Trn1/Trn1n
Neuron Distributed (neuronx-distributed) for Training	pad_model API to pad attention heads that do not divide by the number of NeuronCores, this will allow users to use any supported tensor-parallel degree. See API Reference Guide (neuronx-distributed ) Llama-2-7B model training support (sample script) (tutorial) See more at Neuron Distributed Release Notes (neuronx-distributed) and API Reference Guide (neuronx-distributed )	Trn1/Trn1n
Neuron Distributed (neuronx-distributed) for Inference	T5-3B model inference support (tutorial) pad_model API to pad attention heads that do not divide by the number of NeuronCores, this will allow users to use any supported tensor-parallel degree. See API Reference Guide (neuronx-distributed ) See more at Neuron Distributed Release Notes (neuronx-distributed) and API Reference Guide (neuronx-distributed )	Inf2,Trn1/Trn1n
Transformers Neuron (transformers-neuronx) for Inference	Introducing --model-type=transformer compiler flag that deprecates --model-type=transformer-inference compiler flag. See more at Transformers Neuron (transformers-neuronx) release notes	Inf2, Trn1/Trn1n
PyTorch Neuron (torch-neuronx)	Performance optimizations in torch_neuronx.analyze API. See PyTorch NeuronX Analyze API for Inference Stable Diffusion XL(Refiner and Base) model inference support ( sample script)	Trn1/Trn1n,Inf2
Neuron Compiler (neuronx-cc)	New --optlevel``(or ``-O) compiler option that enables different optimizations with tradeoff between faster model compile time and faster model execution. See more at Neuron Compiler CLI Reference Guide (neuronx-cc) See more at Neuron Compiler (neuronx-cc) release notes	Inf2/Trn1/Trn1n
Neuron Tools	Neuron SysFS support for showing connected devices on trn1.32xl, inf2.24xl and inf2.48xl instances. See Neuron Sysfs User Guide See more at Neuron System Tools	Inf1/Inf2/Trn1/Trn1n
Documentation Updates	Neuron Calculator now supports multiple model configurations for Tensor Parallel Degree computation. See Neuron Calculator Announcement to deprecate --model-type=transformer-inference flag. See Announcing deprecation for --model-type=transformer-inference compiler flag See more at Neuron Documentation Release Notes	Inf1, Inf2, Trn1/Trn1n
Minor enhancements and bug fixes.	See Neuron Components Release Notes	Trn1/Trn1n , Inf2, Inf1
Release Artifacts	see Release Artifacts	Trn1/Trn1n , Inf2, Inf1

For more detailed release notes of the new features and resolved issues, see Neuron Components Release Notes.

To learn about the model architectures currently supported on Inf1, Inf2, Trn1 and Trn1n instances, please see model_architecture_fit.

Neuron 2.13.2 (09/01/2023)

This is a patch release that fixes issues in Kubernetes (K8) deployments related to Neuron Device Plugin crashes and other pod scheduling issues. This release also adds support for zero-based Neuron Device indexing in K8 deployments, see the Neuron K8 release notes for more details on the specific bug fixes.

Updating to latest Neuron Kubernetes components and Neuron Driver is highly encouraged for customers using Kubernetes.

Please follow these instructions in setup guide to upgrade to latest Neuron release.

Neuron 2.13.1 (08/29/2023)

This release adds support for Llama 2 model training (tutorial) using neuronx-nemo-megatron library, and adds support for Llama 2 model inference using transformers-neuronx library (tutorial) .

Please follow these instructions in setup guide to upgrade to latest Neuron release.

Note

Please install transformers-neuronx from https://pip.repos.neuron.amazonaws.com to get latest features and improvements.

This release does not support LLama 2 model with Grouped-Query Attention

Neuron 2.13.0 (08/28/2023)

What’s New

This release introduces support for GPT-NeoX 20B model training in neuronx-distributed including Zero-1 optimizer capability. It also adds support for Stable Diffusion XL and CLIP models inference in torch-neuronx. Neuron 2.13 also introduces AWS Neuron Reference for Nemo Megatron library supporting distributed training of LLMs like GPT-3 175B. This release also introduces other new features, performance optimizations, minor enhancements and bug fixes. This release introduces the following:

What’s New	Details	Instances
AWS Neuron Reference for Nemo Megatron library	Modified versions of the open-source packages NeMo and Apex that have been adapted for use with AWS Neuron and AWS EC2 Trn1 instances. GPT-3 model training support ( tutorial ) See more at neuronx-nemo-megatron github repo	Trn1/Trn1n
Transformers Neuron (transformers-neuronx) for Inference	Latency optimizations for Llama and GPT-2 models inference. Neuron Persistent Cache support (developer guide) See more at Transformers Neuron (transformers-neuronx) release notes	Inf2, Trn1/Trn1n
Neuron Distributed (neuronx-distributed) for Training	Now Stable, removed beta support ZeRO-1 Optimizer support with tensor parallel. (tutorial) Sequence Parallel support. (api guide) GPT-NeoX model training support. (sample script) (tutorial) See more at Neuron Distributed Release Notes (neuronx-distributed) and API Reference Guide (neuronx-distributed )	Trn1/Trn1n
Neuron Distributed (neuronx-distributed) for Inference	KV Cache Support for LLM Inference (release notes)	Inf2,Trn1/Trn1n
PyTorch Neuron (torch-neuronx)	Seedable dropout enabled by default for training KV Cache inference support ( tutorial ) camembert-base training script. (sample script) New models inference support that include Stable Diffusion XL , CLIP (clip-vit-base-patch32 , clip-vit-large-patch14 ) , Vision Perceiver , Language Perceiver and T5	Trn1/Trn1n,Inf2
Neuron Tools	New data types support for Neuron Collective Communication Test Utility (NCCOM-TEST) –check option: fp16, bf16, (u)int8, (u)int16, and (u)int32 Neuron SysFS support for FLOP count(flop_count) and connected Neuron Device ids (connected_devices). See Neuron Sysfs User Guide See more at Neuron System Tools	Inf1/Inf2/Trn1/Trn1n
Neuron Runtime	Runtime version and Capture Time support to NTFF Async DMA copies support to improve Neuron Device copy times for all instance types Logging and error messages improvements for Collectives timeouts and when loading NEFFs. See more at Neuron Runtime Release Notes	Inf1, Inf2, Trn1/Trn1n
End of Support Announcements and Documentation Updates	Announcing End of support for AWS Neuron reference for Megatron-LM starting Neuron 2.13. See more at AWS Neuron reference for Megatron-LM no longer supported Announcing end of support for torch-neuron version 1.9 starting Neuron 2.14. See more at Announcing end of support for torch-neuron version 1.9 Added TensorFlow 2.x (tensorflow-neuronx) analyze_model API section. See more at TensorFlow 2.x (tensorflow-neuron) analyze_model API Upgraded numpy version to 1.21.6 in various training scripts for Text Classification Updated bert-japanese training Script to use multilingual-sentiments dataset. See `hf-bert-jp <https://github.com/aws-neuron/aws-neuron-samples/tree/master/torch-neuronx/training/hf_bert_jp> `_ See more at Neuron Documentation Release Notes	Inf1, Inf2, Trn1/Trn1n
Minor enhancements and bug fixes.	See Neuron Components Release Notes	Trn1/Trn1n , Inf2, Inf1
Known Issues and Limitations	See 2.13.0 Known Issues and Limitations	Trn1/Trn1n , Inf2, Inf1
Release Artifacts	see Release Artifacts	Trn1/Trn1n , Inf2, Inf1

For more detailed release notes of the new features and resolved issues, see Neuron Components Release Notes.

To learn about the model architectures currently supported on Inf1, Inf2, Trn1 and Trn1n instances, please see model_architecture_fit.

2.13.0 Known Issues and Limitations

• Currently we see a NaN generated when the model implementation uses torch.dtype(float32.min) or torch.dtype(float32.max) along with XLA_USE_BF16/XLA_DOWNCAST_BF16. This is because, float32.min or float32.max gets downcasted to Inf in bf16 thereby producing a NaN. Short term fix is that we can use a small/large fp32 number instead of using float32.min/float32.max. Example, for mask creation, we can use -/+1e4 instead of min/max values. The issue will be addressed in future Neuron releases.

Neuron 2.12.2 (08/19/2023)

Patch release to fix a jemalloc conflict for all Neuron customers that use Ubuntu 22. The previous releases shipped with a dependency on jemalloc that may lead to compilation failures in Ubuntu 22 only. Please follow these instructions in setup guide to upgrade to latest Neuron release.

Neuron 2.12.1 (08/09/2023)

Patch release to improve reliability of Neuron Runtime when running applications on memory constrained instances. The Neuron Runtime has reduced the contiguous memory requirement for initializing the Neuron Cores associated with applications. This reduction allows bringup when only small amounts of contiguous memory remain on an instance. Please upgrade to latest Neuron release to use the latest Neuron Runtime.

Neuron 2.12.0 (07/19/2023)

What’s New

This release introduces ZeRO-1 optimizer for model training in torch-neuronx , introduces beta support for GPT-NeoX, BLOOM , Llama and Llama 2(coming soon) models in transformers-neuronx. This release also adds support for model inference serving on Triton Inference Server for Inf2 & Trn1 instances, lazy_load API and async_load API for model loading in torch-neuronx, as well as other new features, performance optimizations, minor enhancements and bug fixes. This release introduces the following:

What’s New	Details	Instances
ZeRO-1 optimizer for model training in torch-neuronx	Support of ZeRO-Stage-1 optimizer ( ZeroRedundancyOptimizer() API) for training models using torch-neuronx See tutorial at ZeRO-1 Tutorial	Inf2, Trn1/Trn1n
Support for new models and Enhancements in transformers-neuronx	[Beta] Support for inference of GPT-NeoX, BLOOM and Llama models. [Beta] Support for Llama 2 coming soon. Please monitor the transformers-neuronx repository for updates. Removed constraints on tp_degree in tensor-parallel configurations for GPT2, OPT, and BLOOM . See more at Transformers Neuron (transformers-neuronx) release notes Added multi-query / multi-group attention support for GPT2. See more at Transformers Neuron (transformers-neuronx) release notes	Inf2, Trn1/Trn1n
Support for Inf2 and Trn1 instances on Triton Inference Server	Support for Model Inference serving on Triton for Inf2 and Trn1 instances. See more at Triton Server Python Backend See tutorial at Triton on SageMaker - Deploying on Inf2	Inf2, Trn1
Support for new computer vision models	Performance optimizations in Stable Diffusion 2.1 model script and added [beta] support for Stable Diffusion 1.5 models. [Beta] Script for training CLIP model for Image Classification. [Beta] Script for inference of Multimodal perceiver model Please check aws-neuron-samples repository	Inf2, Trn1/Trn1n
New Features in neuronx-distributed for training	Added parallel cross entropy loss function. See more at tp_api_guide	Trn1/Trn1n
lazy_load and async_load API for model loading in inference and performance enhancements in torch-neuronx	Added lazy_load and async_load API to accelerate model loading for Inference. See more at PyTorch NeuronX Lazy and Asynchronous Loading API Optimize DataParallel API to load onto multiple cores simultaneously when device IDs specified are consecutive. See more at PyTorch Neuron (torch-neuronx) release notes	Inf2, Trn1/Trn1n
[Beta] Asynchronous Execution support and Enhancements in Neuron Runtime	Added beta asynchronous execution feature which can reduce latency by roughly 12% for training workloads. See more at NeuronX Runtime Configuration AllReduce with All-to-all communication pattern enabled for 16 ranks on TRN1/TRN1N within the instance (intranode) See more at Neuron Runtime Release Notes	Inf1, Inf2, Trn1/Trn1n
Support for distribution_strategy compiler option in neuronx-cc	Support for optional --distribution_strategy compiler option to enable compiler specific optimizations based on distribution strategy used. See more at Neuron Compiler CLI Reference Guide (neuronx-cc)	Inf2, Trn1/Trn1n
New Micro Benchmarking Performance User Guide and Documentation Updates	Added best practices user guide for benchmarking performance of Neuron devices. See more at Benchmarking Guide and Helper scripts Announcing end of support for Ubuntu 18. See more at Announcing end of support for Ubuntu 18 Removed support for Distributed Data Parallel(DDP) Tutorial. Improved sidebar navigation in Documentation. See more at Neuron Documentation Release Notes	Inf1, Inf2, Trn1/Trn1n
Minor enhancements and bug fixes.	See Neuron Components Release Notes	Trn1/Trn1n , Inf2, Inf1
Known Issues and Limitations	See 2.12.0 Known Issues and Limitations	Trn1/Trn1n , Inf2, Inf1
Release Artifacts	see Release Artifacts	Trn1/Trn1n , Inf2, Inf1

For more detailed release notes of the new features and resolved issues, see Neuron Components Release Notes.

To learn about the model architectures currently supported on Inf1, Inf2, Trn1 and Trn1n instances, please see model_architecture_fit.

2.12.0 Known Issues and Limitations

Known Issues in Ubuntu 22 Support

• Several Vision and NLP models on Ubuntu 22 are not supported due to Compilation issues. Issues will be addressed in upcoming releases.

• CustomOp feature failing with seg fault on Ubuntu 22. Issue will be addressed in upcoming releases.

Known issues in certain resnet models on Ubuntu 20

• Known issue with support for resnet-18, resnet-34, resnet-50, resnet-101 and resnet-152 models on Ubuntu 20. Issues will be addressed in upcoming releases.

Neuron 2.11.0 (06/14/2023)

What’s New

This release introduces Neuron Distributed, a new python library to simplify training and inference of large models, improving usability with features like S3 model caching, standalone profiler tool, support for Ubuntu22, as well as other new features, performance optimizations, minor enhancements and bug fixes. This release introduces the following:

What’s New	Details	Instances
New Features and Performance Enhancements in transformers-neuronx	Support for int8 inference. See example at int8 weight storage support Improved prompt context encoding performance. See more at Transformers NeuronX (transformers-neuronx) Developer Guide Improved collective communications performance for Tensor Parallel inference on Inf2 and Trn1. See more at Transformers Neuron (transformers-neuronx) release notes	Inf2, Trn1/Trn1n
Neuron Profiler Tool	Profiling and visualization of model execution on Trainium and Inferentia devices now supported as a stand-alone tool. See more at Neuron Profile User Guide	Inf1, Inf2, Trn1/Trn1n
Neuron Compilation Cache through S3	Support for sharing compiled models across Inf2 and Trn1 nodes through S3 See more at PyTorch NeuronX neuron_parallel_compile CLI	Inf2, Trn1/Trn1n
New script to scan a model for supported/unsupported operators	Script to scan a model for supported/unsupported operators before training, scan output includes supported and unsupported operators at both XLA operators and PyTorch operators level. See a sample tutorial at Analyze for Training Tutorial	Inf2, Trn1/Trn1n
Neuron Distributed Library [Beta]	New Python Library based on PyTorch enabling distributed training and inference of large models. Initial support for tensor-parallelism. See more at NeuronX Distributed	Inf2, Trn1/Trn1n
Neuron Calculator and Documentation Updates	New Neuron Calculator Documentation section to help determine number of Neuron Cores needed for LLM Inference. Added App Note Generative LLM inference with Neuron See more at Neuron Documentation Release Notes	Inf1, Inf2, Trn1/Trn1n
Enhancements to Neuron SysFS	Support for detailed breakdown of memory usage across the NeuronCores See more at Neuron Sysfs User Guide	Inf1, Inf2, Trn1/Trn1n
Support for Ubuntu 22	See more at Setup Guide for setup instructions on Ubuntu22	Inf1, Inf2, Trn1/Trn1n
Minor enhancements and bug fixes.	See Neuron Components Release Notes	Trn1/Trn1n , Inf2, Inf1
Release Artifacts	see Release Artifacts	Trn1/Trn1n , Inf2, Inf1

For more detailed release notes of the new features and resolved issues, see Neuron Components Release Notes.

To learn about the model architectures currently supported on Inf1, Inf2, Trn1 and Trn1n instances, please see model_architecture_fit.

Neuron 2.10.0 (05/01/2023)

What’s New

This release introduces new features, performance optimizations, minor enhancements and bug fixes. This release introduces the following:

What’s New	Details	Instances
Initial support for computer vision models inference	Added Stable Diffusion 2.1 model script for Text to Image Generation Added VGG model script for Image Classification Task Added UNet model script for Image Segmentation Task Please check aws-neuron-samples repository	Inf2, Trn1/Trn1n
Profiling support in PyTorch Neuron(torch-neuronx) for Inference with TensorBoard	See more at Profiling PyTorch Neuron (torch-neuronx) with TensorBoard	Inf2, Trn1/Trn1n
New Features and Performance Enhancements in transformers-neuronx	Support for the HuggingFace generate function. Model Serialization support for GPT2 models. (including model saving, loading, and weight swapping) Improved prompt context encoding performance. See Transformers NeuronX (transformers-neuronx) for examples and usage See more at Transformers Neuron (transformers-neuronx) release notes	Inf2, Trn1/Trn1n
Support models larger than 2GB in TensorFlow 2.x Neuron (tensorflow-neuronx)	See Special Flags for details. (tensorflow-neuronx)	Trn1/Trn1n, Inf2
Support models larger than 2GB in TensorFlow 2.x Neuron (tensorflow-neuron)	See Special Flags for details. (tensorflow-neuron)	Inf1
Performance Enhancements in PyTorch C++ Custom Operators (Beta)	Support for using multiple GPSIMD Cores in Custom C++ Operators See Custom Operators API Reference Guide [Beta]	Trn1/Trn1n
Weight Deduplication Feature (Inf1)	Support for Sharing weights when loading multiple instance versions of the same model on different NeuronCores. See more at NeuronX Runtime Configuration	Inf1
nccom-test - Collective Communication Benchmarking Tool	Supports enabling benchmarking sweeps on various Neuron Collective Communication operations. See NCCOM-TEST User Guide for more details.	Trn1/Trn1n , Inf2
Announcing end of support for tensorflow-neuron 2.7 & mxnet-neuron 1.5 versions	See Announcing end of support for tensorflow-neuron versions 2.7 See Announcing end of support for mxnet-neuron versions 1.5	Inf1
Minor enhancements and bug fixes.	See Neuron Components Release Notes	Trn1/Trn1n , Inf2, Inf1
Release Artifacts	see Release Artifacts	Trn1/Trn1n , Inf2, Inf1

For more detailed release notes of the new features and resolved issues, see Neuron Components Release Notes.

To learn about the model architectures currently supported on Inf1, Inf2, Trn1 and Trn1n instances, please see model_architecture_fit.

Neuron 2.9.1 (04/19/2023)

Minor patch release to add support for deserialized torchscript model compilation and support for multi-node training in EKS. Fixes included in this release are critical to enable training and deploying models with Amazon Sagemaker or Amazon EKS.

Neuron 2.9.0 (03/28/2023)

What’s New

This release adds support for EC2 Trn1n instances, introduces new features, performance optimizations, minor enhancements and bug fixes. This release introduces the following:

What’s New	Details	Instances
Support for EC2 Trn1n instances	Updated Neuron Runtime for Trn1n instances Overall documentation update to include Trn1n instances	Trn1n
New Analyze API in PyTorch Neuron (torch-neuronx)	A new API that return list of supported and unsupported PyTorch operators for a model. See PyTorch NeuronX Analyze API for Inference	Trn1, Inf2
Support models that are larger than 2GB in PyTorch Neuron (torch-neuron) on Inf1	See separate_weights flag to torch_neuron.trace() to support models that are larger than 2GB	Inf1
Performance Improvements	Up to 10% higher throughput when training GPT3 6.7B model on multi-node	Trn1
Dynamic Batching support in TensorFlow 2.x Neuron (tensorflow-neuronx)	See Special Flags for details.	Trn1, Inf2
NeuronPerf support for Trn1/Inf2 instances	Added Trn1/Inf2 support for PyTorch Neuron (torch-neuronx) and TensorFlow 2.x Neuron (tensorflow-neuronx)	Trn1, Inf2
Hierarchical All-Reduce and Reduce-Scatter collective communication	Added support for hierarchical All-Reduce and Reduce-Scatter in Neuron Runtime to enable better scalability of distributed workloads .	Trn1, Inf2
New Tutorials added	Added tutorial to fine-tune T5 model Added tutorial to demonstrate use of Libtorch with PyTorch Neuron (torch-neuronx) for inference [html]	Trn1, Inf2
Minor enhancements and bug fixes.	See Neuron Components Release Notes	Trn1, Inf2, Inf1
Release included packages	see Release Content	Trn1, Inf2, Inf1

For more detailed release notes of the new features and resolved issues, see Neuron Components Release Notes.

To learn about the model architectures currently supported on Inf1, Inf2, Trn1 and Trn1n instances, please see model_architecture_fit.

Neuron 2.8.0 (02/24/2023)

What’s New

This release adds support for EC2 Inf2 instances, introduces initial inference support with TensorFlow 2.x Neuron (tensorflow-neuronx) on Trn1 and Inf2, and introduces minor enhancements and bug fixes.

This release introduces the following:

What’s New	Details
Support for EC2 Inf2 instances	Inference support for Inf2 instances in PyTorch Neuron (torch-neuronx) Inference support for Inf2 instances in TensorFlow 2.x Neuron (tensorflow-neuronx) Overall documentation update to include Inf2 instances
TensorFlow 2.x Neuron (tensorflow-neuronx) support	This releases introduces initial inference support with TensorFlow 2.x Neuron (tensorflow-neuronx) on Trn1 and Inf2
New Neuron GitHub samples	New sample scripts for deploying LLM models with transformer-neuronx under aws-neuron-samples GitHub repository. New sample scripts for deploying models with torch-neuronx under aws-neuron-samples repository GitHub repository.
Minor enhancements and bug fixes.	See Neuron Components Release Notes
Release included packages	see Release Content

For more detailed release notes of the new features and resolved issues, see Neuron Components Release Notes.

Neuron 2.7.0 (02/08/2023)

What’s New

This release introduces new capabilities and libraries, as well as features and tools that improves usability. This release introduces the following:

What’s New	Details
PyTorch 1.13	Support of PyTorch 1.13 version for PyTorch Neuron (torch-neuronx). For resources see PyTorch Neuron
PyTorch DistributedDataParallel (DDP) API	Support of PyTorch DistributedDataParallel (DDP) API in PyTorch Neuron (torch-neuronx). For resources how to use PyTorch DDP API with Neuron, please check neuronx-ddp-tutorial.
Inference support in torch-neuronx	For more details please visit pytorch-neuronx-main` page. You can also try Neuron Inference samples https://github.com/aws-neuron/aws-neuron-samples/tree/master/torch-neuronx in the aws-neuron-samples GitHub repo.
Neuron Custom C++ Operators[Beta]	Initial support for Neuron Custom C++ Operators [Beta] , with Neuron Custom C++ Operators (“CustomOps”) you can now write CustomOps that run on NeuronCore-v2 chips. For more resources please check Neuron Custom C++ Operators [Beta] section.
transformers-neuronx [Beta]	transformers-neuronx is a new library enabling LLM model inference. It contains models that are checkpoint-compatible with HuggingFace Transformers, and currently supports Transformer Decoder models like GPT2, GPT-J and OPT. Please check aws-neuron-samples repository
Neuron sysfs filesystem	Neuron sysfs filesystem exposes Neuron Devices under /sys/devices/virtual/neuron_device providing visibility to Neuron Driver and Runtime at the system level. By performing several simple CLIs such as reading or writing to a sysfs file, you can get information such as Neuron Runtime status, memory usage, Driver info etc. For resources about Neuron sysfs filesystem visit Neuron Sysfs User Guide.
TFLOPS support in Neuron System Tools	Neuron System Tools now also report model actual TFLOPs rate in both neuron-monitor and neuron-top. More details can be found in the Neuron Tools documentation.
New sample scripts for training	This release adds multiple new sample scripts for training models with torch-neuronx, Please check aws-neuron-samples repository
New sample scripts for inference	This release adds multiple new sample scripts for deploying models with torch-neuronx, Please check aws-neuron-samples repository
Neuron GitHub samples repository for Amazon EKS	A new AWS Neuron GitHub samples repository for Amazon EKS, Please check aws-neuron-samples repository

For more detailed release notes of the new features and resolved issues, see Neuron Components Release Notes.

Neuron 2.6.0 (12/12/2022)

This release introduces the support of PyTorch 1.12 version, and introduces PyTorch Neuron (torch-neuronx) profiling through Neuron Plugin for TensorBoard. Pytorch Neuron (torch-neuronx) users can now profile their models through the following TensorBoard views:

• Operator Framework View

• Operator HLO View

• Operator Trace View

This release introduces the support of LAMB optimizer for FP32 mode, and adds support for capturing snapshots of inputs, outputs and graph HLO for debugging.

In addition, this release introduces the support of new operators and resolves issues that improve stability for Trn1 customers.

For more detailed release notes of the new features and resolved issues, see Neuron Components Release Notes.

Neuron 2.5.0 (11/23/2022)

Neuron 2.5.0 is a major release which introduces new features and resolves issues that improve stability for Inf1 customers.

Component	New in this release
PyTorch Neuron (torch-neuron)	PyTorch 1.12 support Python 3.8 support LSTM support on Inf1 R-CNN support on Inf1 Support for new API for core placement Support for improved logging Improved torch_neuron.trace() performance when using large graphs Reduced host memory usage of loaded models in libtorchneuron.so Additional operators support
TensorFlow Neuron (tensorflow-neuron)	tf-neuron-auto-multicore tool to enable automatic data parallel on multiple NeuronCores. Beta support for tracing models larger than 2GB using extract-weights flag (TF2.x only), see TensorFlow 2.x (tensorflow-neuron) Tracing API tfn.auto_multicore Python API to enable automatic data parallel (TF2.x only)

This Neuron release is the last release that will include torch-neuron versions 1.7 and 1.8, and that will include tensorflow-neuron versions 2.5 and 2.6.

In addition, this release introduces changes to the Neuron packaging and installation instructions for Inf1 customers, see Introducing Neuron packaging and installation changes for Inf1 customers for more information.

For more detailed release notes of the new features and resolved issues, see Neuron Components Release Notes.

Neuron 2.4.0 (10/27/2022)

This release introduces new features and resolves issues that improve stability. The release introduces “memory utilization breakdown” feature in both Neuron Monitor and Neuron Top system tools. The release introduces support for “NeuronCore Based Sheduling” capability to the Neuron Kubernetes Scheduler and introduces new operators support in Neuron Compiler and PyTorch Neuron. This release introduces also additional eight (8) samples of models’ fine tuning using PyTorch Neuron. The new samples can be found in the AWS Neuron Samples GitHub repository.

Neuron 2.3.0 (10/10/2022)

This Neuron 2.3.0 release extends Neuron 1.x and adds support for the new AWS Trainium powered Amazon EC2 Trn1 instances. With this release, you can now run deep learning training workloads on Trn1 instances to save training costs by up to 50% over equivalent GPU-based EC2 instances, while getting the highest training performance in AWS cloud for popular NLP models.

What’s New	rn2.3.0_new Introducing Packaging and installation changes Introducing New Neuron GitHub Repositories Announcing Neuron Runtime 1.x (neuron-rtd) end-of-support
Tested workloads and known issues	rn2.3.0_tested rn2.3.0-known-issues

Neural-networks training support

• Supported instances: Trn1

• Supported Frameworks: PyTorch Neuron (torch-neuronx)

• Supported Data-types

  • FP32, BF16

• Supported Rounding Modes

  • Stochastic Rounding (SR)

  • Round Nearest ties to Even (RNE)

• Supported Automatic Casting Methods

  • Neuron automatic casting of FP32 tensors / weights / operations to BF16 - Default mode

  • PyTorch automatic casting

  • Full BF16 automatic casting (via XLA_USE_BF16=1 environment variable)

PyTorch Neuron (torch-neuronx)

• PyTorch 1.11

• Supported instances: Trn1

• Supported Python versions: Python 3.7, Python 3.8

• Eager Debug Mode

• Persistent Cache for compilation

• Collective compute operations: AllReduce

• Optimizers: AdamW, SGD

• Tested loss functions: Negative log-likelihood (NLL), Cross-entropy

• Training Libraries/Frameworks

  • torch.distributed

  • Megatron-LM Reference for Neuron

  • Training Examples

    • Multi-Layer Perceptron Training Tutorial

    • Hugging Face BERT Pretraining Tutorial

    • PyTorch Neuron for Trainium Hugging Face BERT MRPC task finetuning using Hugging Face Trainer API

    • megatron-lm-pretraining-tutorial

• For More information:

  • PyTorch Neuron

  • Developer Guide for Training with PyTorch NeuronX

  • PyTorch Neuron (torch-neuronx) release notes

  • PyTorch Neuron (torch-neuronx) for Training Troubleshooting Guide

Neuron Runtime, Drivers and Networking Components

• Neuron Runtime 2.9

• Supported instances: Trn1, Inf1

• Elastic Fabric Adapter (EFA) @ 800Gbps

• Collective communication operators

  • AllReduce

  • AllGather

  • ReduceScatter

• Release Notes:

  • Neuron Runtime Release Notes

  • Neuron Driver Release Notes

Neuron Tools

• Neuron system tools - Adding Trn1 support to the following tools:

  • neuron-monitor

  • neuron-top

  • neuron-ls

• Release Notes:

  • Neuron System Tools

Developer Flows

• Containers

  • Deep Learning Containers (DLC) supporting PyTorch Neuron (torch-neuronx)

• Multi-Instance distributed workloads orchestration:

  • AWS ParallelCluster (Through custom AMI build)

  • Amazon Elastic Compute Cloud (ECS)

• Supported Amazon Linux Images (AMIs)

  • Ubuntu 20 Neuron DLAMI-base (Python 3.8)

  • Amazon Linux2 Neuron DLAMI-base (Python 3.7)

  • Ubuntu 18 Neuron DLAMI-base (Python 3.7)

  • Ubuntu 18 AMI (Python 3.7)

  • Ubuntu 20 AMI (Python 3.8)

  • Amazon Linux2 AMI (Python 3.7)

The following workloads were tested in this release:

• Distributed data-parallel pre-training of Hugging Face BERT model on single Trn1.32xl instance (32 NeuronCores).

• Distributed data-parallel pre-training of Hugging Face BERT model on multiple Trn1.32xl instances.

• HuggingFace BERT MRPC task finetuning on single NeuronCore or multiple NeuronCores (data-parallel).

• Megatron-LM GPT3 (6.7B parameters) pre-training on single Trn1.32xl instance.

• Megatron-LM GPT3 (6.7B parameters) pre-training on multi Trn1.32xl instances.

• Multi-Layer Perceptron (ML) model training on single NeuronCore or multiple NeuronCores (data-parallel).

• For maximum training performance, please set environment variables XLA_USE_BF16=1 to enable full BF16 and Stochastic Rounding.

This document is relevant for: Inf1, Inf2, Trn1, Trn1n