TriAttention — Efficient long reasoning with trigonometric KV cache compression. Enables OpenClaw local deployment on memory-constrained GPUs.
☆850Aug 4, 2026Updated last month
Alternatives and similar repositories for triattention
Users that are interested in triattention are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- [CVPR2026] Exploring Spatial Intelligence from a Generative Perspective☆32Jun 3, 2026Updated 3 months ago
- [CVPRW 2026 Oral] Less Detail, Better Answers: Degradation-Driven Prompting for VQA☆20Apr 25, 2026Updated 4 months ago
- ☆35Apr 10, 2026Updated 5 months ago
- WorldOlympiad: Can Your World Model Survive a Triathlon?☆56Aug 31, 2026Updated 2 weeks ago
- [ICLR 2026] ParoQuant: Pairwise Rotation Quantization for Efficient Reasoning LLM Inference☆340Aug 19, 2026Updated last month
- Managed Database hosting by DigitalOcean • AdPostgreSQL, MySQL, MongoDB, Kafka, Valkey, and OpenSearch available. Automatically scale up storage and focus on building your apps.
- Frequency-based KV cache pruning for llama.cpp — 25% cache reduction, improved PPL at long context. GPU compaction kernel for HIP/ROCm.☆16Apr 18, 2026Updated 5 months ago
- ☆397Apr 16, 2026Updated 5 months ago
- ☆21Jun 13, 2026Updated 3 months ago
- [CVPR2026] Eliciting Complex Spatial Reasoning in MLLMs through Wide-Baseline Matching☆19Jun 4, 2026Updated 3 months ago
- Unsupervised Learning of Generalizable Robot Motion from Compact State Representation☆45Jun 10, 2026Updated 3 months ago
- DFlash: Block Diffusion for Flash Speculative Decoding☆6,102Aug 18, 2026Updated last month
- Official PyTorch implementation of [PSA: Pyramid Sparse Attention for Efficient Video Understanding and Generation](https://arxiv.org/abs…☆25Jan 25, 2026Updated 7 months ago
- [ICML 2026] World-R1: Reinforcing 3D Constraints for Text-to-Video Generation☆426Jun 3, 2026Updated 3 months ago
- Latent Spatial Memory for Video World Models☆325Aug 27, 2026Updated 3 weeks ago
- Managed Database hosting by DigitalOcean • AdPostgreSQL, MySQL, MongoDB, Kafka, Valkey, and OpenSearch available. Automatically scale up storage and focus on building your apps.
- [ICLR'26] Official PyTorch implementation of "Time Is a Feature: Exploiting Temporal Dynamics in Diffusion Language Models".☆66Mar 5, 2026Updated 6 months ago
- high-performance linear attention kernel library built on TileLang☆704Updated this week
- KV cache compression via block-diagonal rotation. Beats TurboQuant: better PPL (6.91 vs 7.07), 28% faster decode, 5.3x faster prefill, 44…☆1,049Apr 23, 2026Updated 4 months ago
- Long Video Gen Infrastructure☆2,630Sep 7, 2026Updated last week
- [ICML2025] SpargeAttention: A training-free sparse attention that accelerates any model inference.☆1,150Feb 25, 2026Updated 6 months ago
- Quartet II Official Code☆82May 1, 2026Updated 4 months ago
- LLM speculative inference server for heterogeneous hardware & consumer GPUs☆2,868Updated this week
- Dynamic per-token early exit for LLM inference. Skip layers tokens don't need☆33Updated this week
- [ACL'26] EvoToken-DLM (Beyond Hard Masks: Progressive Token Evolution for Diffusion Language)☆49Apr 7, 2026Updated 5 months ago
- Deploy on Railway without the complexity - Free Credits Offer • AdConnect your repo and Railway handles the rest with instant previews. Quickly provision container image services, databases, and storage volumes.
- LLM KV cache compression made easy☆1,208Updated this week
- [ICML 2025] This is the official PyTorch implementation of "ZipAR: Accelerating Auto-regressive Image Generation through Spatial Locality…☆52Mar 25, 2025Updated last year
- Official Implementation of LongLive-RAG: A general retrieval-augmented framework for long video generation.☆110Jun 4, 2026Updated 3 months ago
- Official implementation of "Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decoding"☆1,088May 30, 2026Updated 3 months ago
- [ICLR 2026]QeRL enables RL for 32B LLMs on a single H100 GPU.☆520Mar 30, 2026Updated 5 months ago
- [ICLR2025, ICML2025, NeurIPS2025 Spotlight] Quantized Attention achieves speedup of 2-5x compared to FlashAttention, without losing end-t…☆3,821Jan 17, 2026Updated 8 months ago
- ☆308Apr 5, 2026Updated 5 months ago
- Algorithms for latent compaction☆268Apr 22, 2026Updated 4 months ago
- ☆57May 6, 2026Updated 4 months ago
- Managed Kubernetes at scale on DigitalOcean • AdDigitalOcean Kubernetes includes the control plane, bandwidth allowance, container registry, automatic updates, and more for free.
- Official Implementation of "MemFlow: Flowing Adaptive Memory for Consistent and Efficient Long Video Narratives"☆217Dec 29, 2025Updated 8 months ago
- Accelerating MoE with IO and Tile-aware Optimizations☆769Aug 29, 2026Updated 3 weeks ago
- TokenSpeed is a speed-of-light LLM inference engine.☆2,149Updated this week
- DMax: Aggressive Parallel Decoding for dLLMs☆132Jul 5, 2026Updated 2 months ago
- Autoresearch for GPU kernels. Give it any PyTorch model, go to sleep, wake up to optimized Triton kernels.☆1,561Mar 19, 2026Updated 6 months ago
- A lightweight inference engine supporting speculative speculative decoding (SSD).☆1,004May 10, 2026Updated 4 months ago
- Long-RL: Scaling RL to Long Sequences (NeurIPS 2025)☆729Sep 24, 2025Updated 11 months ago