From-scratch PyTorch implementation of Google's TurboQuant (ICLR 2026) for LLM KV cache compression. 5x compression at 3-bit with 99.5% attention fidelity.
☆1,044Apr 23, 2026Updated 4 months ago
Alternatives and similar repositories for turboquant-pytorch
Users that are interested in turboquant-pytorch are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- TurboQuant: Near-optimal KV cache quantization for LLM inference (3-bit keys, 2-bit values) with Triton kernels + vLLM integration☆1,767Sep 3, 2026Updated last week
- Rust implementation of TurboQuant, PolarQuant, and QJL — zero-overhead vector quantization for semantic search and KV cache compression (…☆26Aug 20, 2026Updated 3 weeks ago
- ☆7,021Jul 20, 2026Updated last month
- ☆200Apr 5, 2026Updated 5 months ago
- KV cache compression via block-diagonal rotation. Beats TurboQuant: better PPL (6.91 vs 7.07), 28% faster decode, 5.3x faster prefill, 44…☆1,047Apr 23, 2026Updated 4 months ago
- GPUs on demand by Runpod - Special Offer Available • AdRun AI, ML, and HPC workloads on powerful cloud GPUs—without limits or wasted spend. Deploy GPUs in under a minute and pay by the second.
- vLLM TurboQuant☆617Jun 25, 2026Updated 2 months ago
- QJL: 1-Bit Quantized JL transform for KV Cache Quantization with Zero Overhead☆99Jan 27, 2025Updated last year
- First open-source implementation of Google TurboQuant (ICLR 2026) -- near-optimal KV cache compression for LLM inference. 5x compression …☆80May 25, 2026Updated 3 months ago
- Port of Facebook's LLaMA model in C/C++☆22Aug 15, 2026Updated 3 weeks ago
- LLM inference in C/C++☆2,374Updated this week
- DFlash: Block Diffusion for Flash Speculative Decoding☆6,076Aug 18, 2026Updated 3 weeks ago
- TriAttention — Efficient long reasoning with trigonometric KV cache compression. Enables OpenClaw local deployment on memory-constrained …☆849Aug 4, 2026Updated last month
- Running a big model on a small laptop☆4,093Mar 19, 2026Updated 5 months ago
- T-Class 2.0: The Standard Framework for AI Sovereignty and Technical Independence☆20Jan 5, 2026Updated 8 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.
- Official Implementation For PolarQuant☆46Apr 2, 2026Updated 5 months ago
- LLM speculative inference server for heterogeneous hardware & consumer GPUs☆2,851Updated this week
- AI agents running research on single-GPU nanochat training automatically☆95,643Mar 26, 2026Updated 5 months ago
- High-Ratio Vector Quantization☆15Feb 3, 2026Updated 7 months ago
- A high-throughput and memory-efficient inference and serving engine for LLMs☆91,571Updated this week
- A unified library of SOTA model optimization techniques like quantization, distillation, pruning, neural architecture search, speculative…☆3,793Updated this week
- Local UI to run and train LLMs and diffusion models. Supports GGUF, MLX, Qwen3.8, DeepSeek-V4, MiniMax-H3, Gemma 4, FLUX and more.☆76,062Updated this week
- A vector index built on TurboQuant, written in Rust with Python bindings☆16,975Aug 21, 2026Updated 3 weeks ago
- LMCache: Supercharge Your LLM with the Fastest KV Cache Layer☆11,770Updated this week
- Managed hosting for WordPress and PHP on Cloudways • AdManaged hosting for WordPress, Magento, Laravel, or PHP apps, on multiple cloud providers. Deploy in minutes on Cloudways by DigitalOcean.
- First MLX implementation of TurboQuant KV cache compression for Apple Silicon☆30Jun 2, 2026Updated 3 months ago
- ☆16Aug 28, 2025Updated last year
- SpectralQuant: Calibrated Eigenbasis Rotation and Water-Filled Bit Allocation for KV-Cache Compression☆202May 15, 2026Updated 3 months ago
- Official inference framework for 1-bit LLMs☆40,231Jul 27, 2026Updated last month
- ☆24Mar 26, 2026Updated 5 months ago
- 🤗 ml-intern: an open-source ML engineer that reads papers, trains models, and ships ML models☆10,814Updated this week
- Autoresearch for GPU kernels. Give it any PyTorch model, go to sleep, wake up to optimized Triton kernels.☆1,555Mar 19, 2026Updated 5 months ago
- Bonsai Demo☆2,288Updated this week
- Kakao Mobility MCP Server for directions and transit information☆11Sep 14, 2025Updated 11 months ago
- Deploy to Railway using AI coding agents - Free Credits Offer • AdUse Claude Code, Codex, OpenCode, and more. Autonomous software development now has the infrastructure to match with Railway.
- A theoretical reconstruction of the Claude Mythos architecture, built from first principles using the available research literature.☆14,896May 23, 2026Updated 3 months ago
- ☆308Apr 5, 2026Updated 5 months ago
- TurboQuant MLX: Flagship high-performance model quantization for Apple Silicon. Squeeze maximum inference speed and memory efficiency out…☆49Aug 5, 2026Updated last month
- TurboQuant reference implementation — KV cache compression with engineering insights (ICLR 2026 paper reproduction)☆17Mar 28, 2026Updated 5 months ago
- LLM inference in C/C++☆127,971Updated this week
- FlashInfer: Kernel Library for LLM Serving☆6,384Updated this week
- SGLang is a high-performance serving framework for large language models and multimodal models.☆35,850Updated this week