Fast LLM speculative inference server for consumer hardware.
☆2,656Jul 13, 2026Updated this week
Alternatives and similar repositories for lucebox
Users that are interested in lucebox are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- Community recipes for serving LLMs on RTX 3090/4090/5090 CUDA gpus. Multi-engine (vLLM, llama.cpp, ik_llama) and model-agnostic. Currentl…☆1,699Updated this week
- DFlash: Block Diffusion for Flash Speculative Decoding☆5,468May 10, 2026Updated 2 months ago
- DFlash & TurboQuant in llama.cpp with up to 3x faster generation and 7.5x more KV cache in same VRAM☆772Updated this week
- LLAMA Turboquant implementation with CUDA support☆687Updated this week
- LLM inference in C/C++☆2,097Updated this week
- Wordpress hosting with auto-scaling - Free Trial Offer • AdFully Managed hosting for WordPress and WooCommerce businesses that need reliable, auto-scalable performance. Cloudways SafeUpdates now available.
- ☆389Apr 16, 2026Updated 2 months ago
- ☆116Apr 28, 2026Updated 2 months ago
- RDNA-native LLM inference engine in Rust.☆482Updated this week
- Pure Rust Inference Engine☆596Updated this week
- llama.cpp fork with TQ3_1S/4S CUDA kernels — 3.5-bit WHT quantization achieving Q4s quality at 10% smaller size. Based on RaBitQ-inspired…☆219Jul 6, 2026Updated last week
- DeepSeek 4 Flash and PRO local inference engine for Metal, CUDA and ROCm☆18,521Jul 3, 2026Updated last week
- Fully uncensored, capability-enhanced abliteration of Qwen3.6-27B. NVFP4 + z-lab DFlash speculative decoding (n=12) on the unified ghcr.i…☆416Jul 3, 2026Updated last week
- Test LLMs on real tasks. Compare models side-by-side.☆373Jun 16, 2026Updated 3 weeks ago
- TokenSpeed is a speed-of-light LLM inference engine.☆1,589Updated this week
- Proton VPN Special Offer - Get 70% off • AdSpecial partner offer. Trusted by over 100 million users worldwide. Tested, Approved and Recommended by Experts.
- A vLLM patch + hand‑written SM120 SASS kernels: 2‑bit MoE experts + an FP4 "delta" cache that recovers precision — matching the official …☆389Updated this week
- Lossless DFlash speculative decoding for MLX on Apple Silicon☆751Jun 11, 2026Updated last month
- ☆6,992Jun 26, 2026Updated 2 weeks ago
- Adaptive Precision for EXpert Models: MoE-aware mixed-precision quantization☆387May 29, 2026Updated last month
- Auto-tuned launcher for GGUF models on llama.cpp / ik_llama.cpp — OpenAI-compatible server with multi-GPU tensor-split, MoE expert placem…☆252Updated this week
- SNDR Core Engine (Genesis) — vLLM runtime patch-overlay for Qwen3.6 + Gemma4 on consumer NVIDIA (Ampere sm_86, 2× A5000/3090). Qwen3.6-35…☆122Updated this week
- llama.cpp fork with TurboQuant WHT-rotated KV cache & weight compression + Gemma 4 MTP and Qwen 3.6 NextN speculative decoding (+30-50% t…☆300Jul 4, 2026Updated last week
- NEW ROCmfp4 format for llama.cpp☆132Jun 13, 2026Updated last month
- Running a big model on a small laptop☆3,980Mar 19, 2026Updated 3 months ago
- Wordpress hosting with auto-scaling - Free Trial Offer • AdFully Managed hosting for WordPress and WooCommerce businesses that need reliable, auto-scalable performance. Cloudways SafeUpdates now available.
- 2.24x decode TPS increase On Qwen 3.6 27B @ temp 0.6 | Native MTP Speculative Decoding On Apple Silicon With No External Drafter.☆1,033Updated this week
- Fast, lossless LLM inference via dual-view diffusion decoding.☆459May 18, 2026Updated last month
- vLLM Qwen 3.6-27B (AWQ-INT4) + DFlash speculative decoding on AMD Strix Halo (gfx1151 iGPU, 128 GB UMA, ROCm 7.13). 24.8 t/s single-strea…☆46May 10, 2026Updated 2 months ago
- AEON vLLM Ultimate — vLLM 0.25.0 built from source for DGX Spark / Blackwell (sm_121a/GB10). One image serves the whole AEON fleet (Gemma…☆71Updated this week
- A pixel art space shooter built entirely by a 9B AI model on a single RTX 3060. Zero hand-written code.☆96Mar 19, 2026Updated 3 months ago
- A vector index built on TurboQuant, written in Rust with Python bindings☆12,690Jun 10, 2026Updated last month
- Qwen3.6-27B on dual RTX 3090 — TP=2 recipe, vLLM nightly, MTP + fp8 KV, validated for concurrent serving☆54Apr 28, 2026Updated 2 months ago
- llama.cpp fork with additional SOTA quants and improved performance☆2,892Updated this week
- KVarN is a native vLLM KV-cache quantization backend for your agents: 3-5x more context, throughput above FP16, and FP16-level accuracy. …☆437Jun 22, 2026Updated 3 weeks ago
- End-to-end encrypted email - Proton Mail • AdSpecial offer: 40% Off Yearly / 80% Off First Month. All Proton services are open source and independently audited for security.
- ROCmFPX Family for AMD Hardware and Processors. More quants and special agent quants☆107Updated this week
- Run GLM-5.2 (744B MoE) on a 25GB-RAM consumer machine — pure C, zero deps, experts streamed from disk. Tiny engine, immense model. 🐦☆11,798Updated this week
- One-click Qwen3.6-27B inference on Windows. 158 tok/s on RTX 5090, 72 tok/s on RTX 3090. Native, no WSL, no Docker, no telemetry.☆222May 14, 2026Updated 2 months ago
- The agent that grows with you☆214,747Updated this week
- DeepSpec: a full-stack codebase for training and evaluating speculative decoding algorithms☆6,638Updated this week
- Exact speculative decoding on Apple Silicon, powered by MLX.☆380Apr 20, 2026Updated 2 months ago
- Sequential Monte Carlo Speculative Decoding☆52Updated this week