Pinned Loading

1. trainium-collectives trainium-collectives Public

   Predict and compare all-reduce performance across NeuronLink and EFA with explainable ring, tree, and hierarchical cost models.

   Python

2. vlm-inference-profiler vlm-inference-profiler Public

   Per-component latency and VRAM profiling of Qwen2.5-VL-7B across modality-selective quantization configs on T4 GPU

   Jupyter Notebook

3. aws-nki aws-nki Public

   Forked from aws-neuron/nki-samples

   Python

4. clip-neuron-profiling clip-neuron-profiling Public

   Profiling CLIP VLM on AWS Inferentia2: What the Execution Timeline Reveals About Multi-Modal Bottlenecks

   Python 1

5. ooink-ai ooink-ai Public

   An interactive Pig AI assistant that answers customer questions like a host deployed as a tablet kiosk at Ooink Ramen Restaurant in Seattle

   Dart

6. tej-stack tej-stack Public

   Evidence-first UX and product-quality audits for completed features

   Shell