fla-org / flameLinks
π₯ A minimal training framework for scaling FLA models
β146Updated 3 weeks ago
Alternatives and similar repositories for flame
Users that are interested in flame are comparing it to the libraries listed below
Sorting:
- Efficient triton implementation of Native Sparse Attention.β155Updated last week
- XAttention: Block Sparse Attention with Antidiagonal Scoringβ158Updated 3 weeks ago
- Triton-based implementation of Sparse Mixture of Experts.β216Updated 6 months ago
- [ICLR 2025] COAT: Compressing Optimizer States and Activation for Memory-Efficient FP8 Trainingβ203Updated 2 weeks ago
- β74Updated 3 months ago
- β129Updated 3 months ago
- β93Updated 2 weeks ago
- β53Updated this week
- [ICML 2024] Quest: Query-Aware Sparsity for Efficient Long-Context LLM Inferenceβ291Updated 6 months ago
- Code for paper: [ICLR2025 Oral] FlexPrefill: A Context-Aware Sparse Attention Mechanism for Efficient Long-Sequence Inferenceβ107Updated 2 weeks ago
- PyTorch bindings for CUTLASS grouped GEMM.β93Updated this week
- A sparse attention kernel supporting mix sparse patternsβ219Updated 3 months ago
- [ICLR2025] Codebase for "ReMoE: Fully Differentiable Mixture-of-Experts with ReLU Routing", built on Megatron-LM.β74Updated 5 months ago
- SeerAttention: Learning Intrinsic Sparse Attention in Your LLMsβ102Updated this week
- Triton implementation of FlashAttention2 that adds Custom Masks.β117Updated 9 months ago
- β83Updated last month
- Flash-Muon: An Efficient Implementation of Muon Optimizerβ115Updated this week
- 16-fold memory access reduction with nearly no lossβ94Updated 2 months ago
- Official repository for DistFlashAttn: Distributed Memory-efficient Attention for Long-context LLMs Trainingβ208Updated 9 months ago
- Low-bit optimizers for PyTorchβ128Updated last year
- β248Updated last year
- β78Updated this week
- π Efficiently (pre)training foundation models with native PyTorch features, including FSDP for training and SDPA implementation of Flashβ¦β249Updated this week
- β144Updated 2 years ago
- [ICML 2025] Fourier Position Embedding: Enhancing Attentionβs Periodic Extension for Length Generalizationβ69Updated 4 months ago
- Tritonbench is a collection of PyTorch custom operators with example inputs to measure their performance.β127Updated this week
- GEAR: An Efficient KV Cache Compression Recipefor Near-Lossless Generative Inference of LLMβ163Updated 10 months ago
- β105Updated 9 months ago
- Code for exploring Based models from "Simple linear attention language models balance the recall-throughput tradeoff"β234Updated 3 months ago
- Fast and memory-efficient exact attentionβ68Updated 3 months ago