Community recipes for serving LLMs on RTX 3090/4090/5090 CUDA gpus. Multi-engine (vLLM, sglang, llama.cpp and other custom engines) and model-agnostic. Currently shipping Qwen3.8-27B Qwen3.6 35B Gemma 4 26B Gemma 4 31B configs for 1×/2×/4x/8x cards.
☆2,431Oct 22, 2026Updated this week
Alternatives and similar repositories for club-3090
Users that are interested in club-3090 are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- ☆119Apr 28, 2026Updated 5 months ago
- LLM speculative inference server for heterogeneous hardware & consumer GPUs☆2,942Updated this week
- Serve large Qwen models fast on the GPUs you actually own. Qwen3.8-27B on a single 24 GB card with vLLM: 127 tok/s single-user (381 when …☆2,154Updated this week
- Hyper optimised Qwen3.8-27B inference on one RTX 3090: ReplaySSM, MTP3, reasoning effort, C1-C8 batching, and native Windows and Linux bu…☆465Oct 16, 2026Updated last 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…☆133Sep 30, 2026Updated 3 weeks 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.
- Qwen3.6-27B on dual RTX 3090 — TP=2 recipe, vLLM nightly, MTP + fp8 KV, validated for concurrent serving☆58Apr 28, 2026Updated 5 months ago
- Single-file installer for a club-3090 webserver providing an admin control panel, a reverse proxy that automatically routes requests to t…☆28Jul 8, 2026Updated 3 months ago
- KVarN, KV cache precision tail, low-bit quants in llama.cpp for longer context of better precision in the same VRAM☆1,168Updated this week
- Unlock P2P comms between consumer NVIDIA GPUs☆599Sep 16, 2026Updated last month
- llama.cpp fork for significantly improved performance on Ampere (especially RTX 3090 / 3090 Ti): example: 95+ tok/s over a 100K-token gen…☆185Updated this week
- Reliable model swapping for any local OpenAI/Anthropic compatible server - llama.cpp, vllm, etc☆6,051Updated this week
- Qwen3.8-Flash-Next on 2× RTX 3090: with 128 GB RAM up to 4,191 tok/s prefill · 111.5 tok/s decode (131K prompt), full 256K window at 2,86…☆124Oct 15, 2026Updated last week
- High-performance single-GPU inference for selected model checkpoints and GPUs.☆3,125Updated this week
- llama.cpp fork with additional SOTA quants and improved performance☆3,321Updated this week
- Serverless GPU API endpoints on Runpod - Get Bonus Credits • AdSkip the infrastructure headaches. Auto-scaling, pay-as-you-go, no-ops approach lets you focus on innovating your application.
- Experimental llama.cpp fork for inference research and development☆888Updated this week
- vLLM Docker Container for the latest Qwen 27b☆52Sep 27, 2026Updated 3 weeks ago
- Qwen3.8-Flash-Next on any consumer hardware: one-click install for Windows / Linux. Strata inference engine, OpenAI/Anthropic API on loca…☆30,418Updated this week
- An optimized quantization and inference library for running LLMs locally on modern consumer-class GPUs☆1,729Updated this week
- llama.cpp/ik_llama.cpp launcher: loads big MoE models across mismatched multi-GPU rigs by exact VRAM math.☆282Updated this week
- A harness optimized to smaller LLMs☆2,698Sep 18, 2026Updated last month
- RTX 6000 Pro Wiki — Running Large LLMs (Qwen3.5-397B, Kimi-K2.5, GLM-5) on PCIe GPUs without NVLink☆1,148Updated 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.☆231May 14, 2026Updated 5 months ago
- CLI port of BenchLocal quality bench packs — runs LLM behavioral evals against any OpenAI-compatible endpoint. Companion to club-3090.☆18Oct 16, 2026Updated last week
- 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 vLLM patch + hand‑written SM120 SASS kernels: 2‑bit MoE experts + an FP4 "delta" cache that recovers precision — matching the official …☆545Updated this week
- An all-in-one, pure C++ inference engine for audio models, powered by ggml. Supports TTS, STT, VAD, voice conversion, music generation, a…☆3,682Updated this week
- Project Maya - run GLM-5.3-Flash (321B MoE) and DeepSeek-V4.1-Flash-1017 (748B MoE) privately on your own GPU(s), from one card to 16, NV…☆1,458Updated this week
- Reproducible llama.cpp configs + per-category quality benches for Qwen3.6-27B on a single RTX 4090. Winners, dead ends, and the silent-co…☆24Apr 26, 2026Updated 5 months ago
- NInfer SM80 port for NVIDIA CMP 170HX☆28Aug 22, 2026Updated 2 months ago
- Deployment kit: Qwen3.8-27B EXL3 3.5bpw target + DFlash2 EXL3 5.0bpw speculative draft — launcher, config, model cards☆232Sep 1, 2026Updated last month
- LLM inference in C/C++☆2,435Updated this week
- Hand-written NVFP4 W4A16 CUDA kernels for Volta☆175Aug 19, 2026Updated 2 months ago
- Core, junction, and VRAM temps for GDDR6, GDDR6X, GDDR7 NVIDIA GPUs in Linux☆100Sep 24, 2026Updated 3 weeks 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.
- KVarN is a native vLLM KV-cache quantization backend for your agents: 3-5x more context, throughput above FP16, and FP16-level accuracy. …☆511Jun 22, 2026Updated 4 months ago
- AI coding agent optimized for small LLMs. 87% benchmark with 4B-active model.☆2,029Aug 12, 2026Updated 2 months ago
- DFlash: Block Diffusion for Flash Speculative Decoding☆6,190Aug 18, 2026Updated 2 months ago
- A tool to unlobotomize your NVIDIA card!☆782Oct 8, 2026Updated 2 weeks ago
- Control panel for VLLM, Sglang, llama.cpp, exllamav3☆1,834Oct 12, 2026Updated last week
- Qwen3.8-27B on 16-32 GB Nvidia GPUs one-click install for Windows / Linux☆591Oct 9, 2026Updated 2 weeks ago
- Fastest validated sglang recipe for Qwen3.8-27B (AWQ-INT4 + DSpark) on RTX 3090s — 1×/2×/4× GPU presets, image+video, full benchmarks☆64Updated this week