yanring / Megatron-MoE-ModelZooLinks
Best practices for testing advanced Mixtral, DeepSeek, and Qwen series MoE models using Megatron Core MoE.
☆21Updated 2 weeks ago
Alternatives and similar repositories for Megatron-MoE-ModelZoo
Users that are interested in Megatron-MoE-ModelZoo are comparing it to the libraries listed below
Sorting:
- Odysseus: Playground of LLM Sequence Parallelism☆70Updated last year
- ☆114Updated 3 weeks ago
- PyTorch bindings for CUTLASS grouped GEMM.☆99Updated 3 weeks ago
- ☆77Updated 2 months ago
- Boosting 4-bit inference kernels with 2:4 Sparsity☆79Updated 9 months ago
- Accelerate LLM preference tuning via prefix sharing with a single line of code☆41Updated last month
- Official repository for DistFlashAttn: Distributed Memory-efficient Attention for Long-context LLMs Training☆210Updated 10 months ago
- ☆49Updated last month
- Vocabulary Parallelism☆19Updated 3 months ago
- ☆105Updated 10 months ago
- ☆63Updated this week
- [ICLR2025] Breaking Throughput-Latency Trade-off for Long Sequences with Speculative Decoding☆116Updated 6 months ago
- DeeperGEMM: crazy optimized version☆69Updated last month
- Quantized Attention on GPU☆44Updated 7 months ago
- ☆39Updated last year
- ☆81Updated 7 months ago
- (NeurIPS 2022) Automatically finding good model-parallel strategies, especially for complex models and clusters.☆39Updated 2 years ago
- Estimate MFU for DeepSeekV3☆24Updated 5 months ago
- The official implementation of paper: SimLayerKV: A Simple Framework for Layer-Level KV Cache Reduction.☆46Updated 8 months ago
- ☆74Updated 4 years ago
- Transformers components but in Triton☆34Updated last month
- 16-fold memory access reduction with nearly no loss☆99Updated 3 months ago
- Tritonbench is a collection of PyTorch custom operators with example inputs to measure their performance.☆136Updated this week
- ☆86Updated 3 months ago
- ☆26Updated last year
- ☆21Updated 3 months ago
- ☆30Updated last month
- GPTQ inference TVM kernel☆40Updated last year
- ☆60Updated 2 months ago
- ☆84Updated 3 years ago