Code for the paper *Attention Drift: What Speculative Decoding Models Learn*.
☆33May 12, 2026Updated 5 months ago
Alternatives and similar repositories for Attention-Drift
Users that are interested in Attention-Drift are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- SGLang fork to reproduce the official Mirai Labs DFlash-TfM results☆24Oct 14, 2026Updated last week
- Official implementation of "Learning To Draft: Adaptive Speculative Decoding with Reinforcement Learning" (ICLR 2026)☆23Mar 1, 2026Updated 7 months ago
- The official repo of "Draft-OPD: On-Policy Distillation for Speculative Draft Models"☆83Sep 4, 2026Updated last month
- Open, calibrated, single-pass typed-decision models (4B & 27B) for finance and trading☆41Oct 1, 2026Updated 3 weeks ago
- ☆22May 22, 2026Updated 5 months ago
- GPU virtual machines on DigitalOcean Gradient AI • AdGet to production fast with high-performance AMD and NVIDIA GPUs you can spin up in seconds. The definition of operational simplicity.
- Open reproduction of TypeSafe Jev: a 150M typed decision engine (noul/choice/score in one non-autoregressive pass, calibrated confidence)…☆46Oct 6, 2026Updated 2 weeks ago
- ☆16Oct 2, 2026Updated 3 weeks ago
- ☆22Oct 14, 2024Updated 2 years ago
- PARD: Accelerating LLM Inference with Low-Cost PARallel Draft Model Adaptation (ICLR 26)☆37Jun 10, 2026Updated 4 months ago
- A PyTorch native library for training speculative decoding models☆266Updated this week
- Official Implementation of "Learning Harmonized Representations for Speculative Sampling" (HASS)☆56Mar 14, 2025Updated last year
- One RGB camera, 8.2 mm from motion capture - and no idea how big the room is. Reproducible monocular SfM on TUM fr3, plus the tool that s…☆25Aug 29, 2026Updated last month
- [EMNLP2026 Main] Official implementation of “Domino: Decoupling Causal Modeling from Autoregressive Drafting in Speculative Decoding”.☆140Jul 25, 2026Updated 2 months ago
- Ulysses sequence-parallel all-to-all as a torch custom op, moved by the GPU copy engines into torch symmetric memory. Zero SM usage; 1.66…☆54Aug 25, 2026Updated last month
- 1-Click AI Models by DigitalOcean Gradient • AdDeploy popular AI models on DigitalOcean Gradient GPU virtual machines with just a single click. Zero configuration with optimized deployments.
- HALO: Hadamard-Assisted Low-Precision Optimization and Training method for finetuning LLMs. 🚀 The official implementation of https://arx…☆32Feb 17, 2025Updated last year
- Unlocking Lossless Speedups in LLMs via Discrete Diffusion☆106Oct 12, 2026Updated last week
- Implemented and benchmarked KV cache compression methods for LLM inference in Triton. Optimized kernels for KIVI and TurboQuant with Flas…☆18Jun 19, 2026Updated 4 months ago
- 🔥 [ICML'26] ParisKV: Fast and Drift-Robust KV-Cache Retrieval for Long-Context LLMs☆36Aug 14, 2026Updated 2 months ago
- ☆32Oct 16, 2026Updated last week
- Repository of the paper "On the Trade-off between Over-smoothing and Over-squashing in Deep Graph Neural Networks" published in ACM CIKM …☆19Aug 8, 2023Updated 3 years ago
- [NeurIPS 2026] Free Draft-and-Verification: Toward Lossless Parallel Decoding for Diffusion Large Language Models☆24Sep 29, 2026Updated 3 weeks ago
- ☆41Aug 16, 2026Updated 2 months ago
- Quasar Attention Continuous-Time Memory Dynamics for Scaling Models Context to Millions of Tokens☆29Mar 26, 2026Updated 6 months ago
- AI Agents on DigitalOcean Gradient AI Platform • AdBuild production-ready AI agents using customizable tools or access multiple LLMs through a single endpoint. Create custom knowledge bases or connect external data.
- A small Node.js script that polls a website and sends a notification when there's a change. Inspired by my attempts to get a close hotel …☆13Nov 10, 2022Updated 3 years ago
- (IJCV 2023) Offical implementation of "SCT: A Simple Baseline for Parameter-Efficient Fine-Tuning via Salient Channels"☆13Mar 20, 2025Updated last year
- みんなのローカルLLMベンチマーク結果をあつめるレポジトリ☆35Updated this week
- A small (~1000LoC) embeddable LISP interpreter☆13Feb 28, 2025Updated last year
- Android向けに実装された、エンジンにmozcを採用している日本語IME☆29Sep 17, 2026Updated last month
- GEMMs with metal☆26Oct 5, 2026Updated 2 weeks ago
- Seeing is Not Reasoning: MVPBench for Graph-based Evaluation of Multi-path Visual Physical CoT☆14Jul 30, 2025Updated last year
- Official vLLM evaluation for OLMo and NCP-ArchPreview: GSM8K, SciQ, Core88, RULER, and HELMET☆27Sep 11, 2026Updated last month
- 亚微秒级 System-1 模型，准确率服从高斯分布☆32Sep 21, 2026Updated last month
- Simple, predictable pricing with DigitalOcean hosting • AdAlways know what you'll pay with monthly caps and flat pricing. Enterprise-grade infrastructure trusted by 600k+ customers.
- T5 model optimized with FlexAttention☆14Sep 30, 2024Updated 2 years ago
- Breakable CUDA graph capture for PyTorch - capture compatible regions as CUDA graphs while running incompatible operations eagerly betwee…☆29Oct 8, 2026Updated 2 weeks ago
- An extention to the GaLore paper, to perform Natural Gradient Descent in low rank subspace☆19Oct 21, 2024Updated 2 years ago
- [ICML 2026] Jacobi Forcing: Fast and Accurate Diffusion-style Decoding☆118Feb 20, 2026Updated 8 months ago
- Run oversized sparse MoE GGUF models with bounded NVMe-to-RAM-to-VRAM caching, native multi-GPU planning, and a Linux desktop Studio.☆20Oct 12, 2026Updated last week
- ☆34Sep 14, 2026Updated last month
- Open-source Human Feedback Library☆11Oct 25, 2023Updated 2 years ago