Skip to content

Powering AWS purpose-built machine learning chips. Blazing fast and cost effective, natively integrated into PyTorch and TensorFlow and integrated with your favorite AWS services

active 2023-08-152026-08-18 (UTC)

Complete coverage26,458 / 26,458 hourly files (100%) · 2 absent upstream2023-08-152026-08-21 (UTC)
Events
3.1K
Pushes
217
Pull requests
240
Issues
695
Stars
223
Forks
78

Activity over time

Daily event counts in the loaded window

Line chart, 1100 days from 2023-08-15 to 2026-08-18. Pushes: 217 total, peak 7 in a day. Pull requests: 240 total, peak 10 in a day. Issues: 695 total, peak 26 in a day. Comments: 1,318 total, peak 53 in a day. Stars: 223 total, peak 3 in a day.

  • Pushes
  • Pull requests
  • Issues
  • Comments
  • Stars

Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.

Top contributors

Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity

ContributorContributionsPushesPRsComments
jeffhataws832251
aws-taylor800054
aws-donkrets750058
natemail-aws6923440
erickson-doug6516913
mrnikwaws600041
jluntamazon530048
aws-rhsoln510046
ivashkst5123195
awsjoshir493880
micwade-aws446814
musunita441735
lroyprad4318240
JingyaHuang430023
eshalakhotia41111
xanderdunn410020
jyang-aws410027
aws-rxgupta3614220
nandeeka350017
AWSNB320031

Recent activity

Latest issues, pull requests and releases

  • Issue#1354github-actions[bot]2026-06-27 01:59
    Using dynamic_range loops with tensor_copy can cause errors with no information
  • Issue comment#1354github-actions[bot]2026-06-27 01:59
    Using dynamic_range loops with tensor_copy can cause errors with no information
  • Issue#1354natetyoung2026-06-27 01:59
    Using dynamic_range loops with tensor_copy can cause errors with no information
  • Issue comment#1338geetasg2026-05-29 21:40
    Inf2 ERROR NRT:nrt_allocate_neuron_cores
  • Issue#1300mehvaibh2026-05-22 18:40
    Neuron scheduler extension initializes device usage map as [true] after operator uninstall/reinstall
  • Issue#1329github-actions[bot]2026-05-17 14:38
    neuronx-cc OOM on pure-PyTorch gather+SDPA attention at FlashVSR scale (inf2.8xlarge 128GB host RAM)
  • Issue#1327Mathias52026-05-17 14:37
    NKI 0.2.0 toolchain: published nki-samples attention examples (v1, v2) fail at trace time on current Neuron DLAMI
  • Issue#1326github-actions[bot]2026-05-14 06:02
    [Bug] NCC_IBIR243 Access pattern out of bounds when compiling EVA-ViT-G for inf2
  • Issue comment#1326github-actions[bot]2026-05-14 06:02
    [Bug] NCC_IBIR243 Access pattern out of bounds when compiling EVA-ViT-G for inf2
  • Issue#1316aws-viekash2026-05-13 23:13
    [torch-xla 2.9] Training without torch_xla.sync() silently produces no weight updates -- silent correctness failure
  • Issue comment#1316aws-viekash2026-05-13 23:13
    [torch-xla 2.9] Training without torch_xla.sync() silently produces no weight updates -- silent correctness failure
  • Issue comment#1323github-actions[bot]2026-05-08 16:27
    yolo26l on inf2/trn1 (NCv2): numerical error in C2PSA Attention.forward + LegalizeSundaMacro internal error at bs>=4
  • Issue#1281musunita2026-05-01 23:53
    NKI 2 DMA Copy in/out shape must match Errors
  • Pull request#1319wypku2026-04-25 08:00
  • Issue comment#1308mharwani2026-04-24 23:56
    nl.multiply and related ops on uint32 tiles lose precision above 2^24 via float32 activation path
  • Issue#1308mharwani2026-04-24 23:56
    nl.multiply and related ops on uint32 tiles lose precision above 2^24 via float32 activation path
  • Issue#1318github-actions[bot]2026-04-23 03:21
    [torch-xla 2.9] bf16 model + fp32 intermediate arithmetic causes dtype mismatch crash after XLA_USE_BF16 migration
  • Issue#1311musunita2026-04-17 21:35
    NKI: trn1 verifier rejects shared-memory instruction in fused XLA graph spanning ao_to_mo_transform + mp2_energy
  • Issue comment#1310github-actions[bot]2026-04-17 18:41
    Add support for PJRT_Client_CreateUninitializedBuffer
  • Issue#1310steeve2026-04-17 18:41
    Add support for PJRT_Client_CreateUninitializedBuffer
  • Issue#1308musunita2026-04-15 22:27
    nl.multiply and related ops on uint32 tiles lose precision above 2^24 via float32 activation path
  • Issue comment#1289rohis06-aws2026-04-15 20:50
    [Bug] GQA sharding broken for TP < num_kv_heads in neuronx-distributed-inference 0.8.x
  • Issue#1308github-actions[bot]2026-04-15 01:19
    nl.multiply and related ops on uint32 tiles lose precision above 2^24 via float32 activation path
  • Issue#1308github-actions[bot]2026-04-15 01:19
    nl.multiply and related ops on uint32 tiles lose precision above 2^24 via float32 activation path
  • Issue comment#1307github-actions[bot]2026-04-13 23:36
    VSCode Extension

Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 223 stars here means stars gained during the window, not the repo's star count.