Pinned Loading

1. Kimi-Linear Kimi-Linear Public

   Forked from MoonshotAI/Kimi-Linear

   Up to 6x decoding throughput improvement for context as long as 1M tokens, reduction of KV cache by up to 75%

   1

2. TensorScope TensorScope Public

   Easily benchmark training of any model by (op type+parameters) to spot real bottlenecks. Catches more details then native tfprof, especially for RNN/LSTM models

   HTML 1

3. fast-pixel-classification-dnn-caffe fast-pixel-classification-dnn-caffe Public

   Real-time detection of roads, cars, pedestrians

   Python 3 2

4. fast-cityscapes fast-cityscapes Public

   Evaluate a deep neural network on CityScapes dataset

   Python

5. DeepSpeed DeepSpeed Public

   Forked from deepspeedai/DeepSpeed

   DeepSpeed is a deep learning optimization library that makes distributed training easy, efficient, and effective.

   Python 1

6. DeepSpeedExamples DeepSpeedExamples Public

   Forked from deepspeedai/DeepSpeedExamples

   Example models using DeepSpeed

   Python 1