TorchNeuron Space lists which Hub models run unmodified on AWS Trainium
A new Space called torch-neuronx model support lets users check whether any model on the Hub can run unmodified on AWS Trainium. The interface shows results for the most downloaded and trending models, indicating if the model passes, how many Trainium cores it is sharded over, wall‑time including compilation, and peak device memory. Users can switch between Selected, Trending or Downloads views…
Key points
- Space shows pass/fail, core count, wall‑time and memory for each Hub model on Trainium
- Only Transformers models for text‑generation and image‑text‑to‑text are tested, up to 4 Trainium2 devices (16 cores)
- Run script compiles with torch.compile(backend="neuron") and generates 32 tokens for validation
At the time of writing the Space only tests Transformers models for text‑generation and image‑text‑to‑text tasks, and limits testing to models that fit on four Trainium2 devices – a maximum of 16 cores. Tests run on a trn2 instance in the AWS Neuron Deep Learning Container, compile the model with torch.compile(backend="neuron"), and generate 32 tokens for validation. The Space also provides the exact Neuron stack versions, required packages and the run.py script used, plus commands to reuse compiled graphs from the public NEFF cache bucket.
The story so far
5 episodes →- TorchNeuron Space lists which Hub models run unmodified on AWS Trainiumthis story
Does Your Model Run on AWS Trainium? Look It Up
Hugging Face · 23 October 2026
Loading the full article…
This text was published by Hugging Face and written by David Corvoysier. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Generative AI & Models
All →- Deepseek launches V4.1 Flash 1017 model with 1 M token context · 1 src
- OpenAI says its AI agents weakened internal misalignment monitoring system · 33 src
- Opinion: Jev’s hype cycle shows AI marketing overpromises and fizzles · 1 src
- OpenAI brings GPT-6 Astra, GPT-6.1 Sol, GPT-6 Luna and Codex to Google Cloud · 20 src
- Tencent releases WeVisOne-VL 4B vision-language model on Hugging Face · 1 src
Comments
via GitHub Discussions