hao-ai-lab / DistCALinks
Efficient Long-context Language Model Training by Core Attention Disaggregation
☆85Updated this week
Alternatives and similar repositories for DistCA
Users that are interested in DistCA are comparing it to the libraries listed below
Sorting:
- ☆117Updated 8 months ago
- ☆52Updated 8 months ago
- [Archived] For the latest updates and community contribution, please visit: https://github.com/Ascend/TransferQueue or https://gitcode.co…☆13Updated 2 weeks ago
- High-performance distributed data shuffling (all-to-all) library for MoE training and inference☆109Updated last month
- DeeperGEMM: crazy optimized version☆73Updated 8 months ago
- ☆38Updated 5 months ago
- ☆65Updated 9 months ago
- 🔥 LLM-powered GPU kernel synthesis: Train models to convert PyTorch ops into optimized Triton kernels via SFT+RL. Multi-turn compilation…☆116Updated 2 months ago
- Distributed MoE in a Single Kernel [NeurIPS '25]☆190Updated this week
- Accelerate LLM preference tuning via prefix sharing with a single line of code☆51Updated 6 months ago
- A Suite for Parallel Inference of Diffusion Transformers (DiTs) on multi-GPU Clusters☆55Updated last year
- An experimental communicating attention kernel based on DeepEP.☆35Updated 6 months ago
- Autonomous GPU Kernel Generation via Deep Agents☆223Updated this week
- Building the Virtuous Cycle for AI-driven LLM Systems☆140Updated this week
- A NCCL extension library, designed to efficiently offload GPU memory allocated by the NCCL communication library.☆87Updated last month
- Vortex: A Flexible and Efficient Sparse Attention Framework☆45Updated last week
- [ICLR2025] Breaking Throughput-Latency Trade-off for Long Sequences with Speculative Decoding☆137Updated last year
- ☆132Updated 8 months ago
- [NeurIPS 2025] Scaling Speculative Decoding with Lookahead Reasoning☆62Updated 3 months ago
- ☆221Updated 2 months ago
- Quantized Attention on GPU☆44Updated last year
- DeepSeek-V3.2-Exp DSA Warmup Lightning Indexer training operator based on tilelang☆43Updated 2 months ago
- ☆129Updated 5 months ago
- 16-fold memory access reduction with nearly no loss☆109Updated 10 months ago
- Debug print operator for cudagraph debugging☆14Updated last year
- ☆32Updated 7 months ago
- Odysseus: Playground of LLM Sequence Parallelism☆79Updated last year
- [ASPLOS'26] Taming the Long-Tail: Efficient Reasoning RL Training with Adaptive Drafter☆130Updated last month
- Fast and memory-efficient exact attention☆18Updated last week
- Framework to reduce autotune overhead to zero for well known deployments.☆94Updated 4 months ago