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.
☆2,497Nov 11, 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:
- 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 …☆2,679Updated 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…☆512Oct 31, 2026Updated last week
- ☆119Apr 28, 2026Updated 6 months ago
- Lucebox inference engine for heterogeneous hardware & consumer GPUs☆3,003Updated 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…☆133Sep 30, 2026Updated last month
- 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.
- Qwen3.6-27B on dual RTX 3090 — TP=2 recipe, vLLM nightly, MTP + fp8 KV, validated for concurrent serving☆58Apr 28, 2026Updated 6 months ago
- 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…☆141Oct 15, 2026Updated 3 weeks ago
- Unlock P2P comms between consumer NVIDIA GPUs☆719Oct 26, 2026Updated 2 weeks 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 4 months ago
- KVarN, KV cache precision tail, low-bit quants in llama.cpp for longer context of better precision in the same VRAM☆1,193Updated this week
- Source-available C++/CUDA inference engine for Qwen3.8-27B INT4 on 3x RTX 3090☆49Updated this week
- llama.cpp fork for significantly improved performance on Ampere generation Nvidia cards (especially RTX 3090 / 3090 Ti): example: 110+ to…☆371Updated this week
- Reliable model swapping for any local AI server. Chosen by thousands of users. Batteries included. If we don't have it, you probably don…☆6,241Updated this week
- CLI port of BenchLocal quality bench packs — runs LLM behavioral evals against any OpenAI-compatible endpoint. Companion to club-3090.☆20Nov 4, 2026Updated last 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.
- Experimental llama.cpp fork for inference research and development☆907Updated this week
- llama.cpp fork with additional SOTA quants and improved performance☆3,387Updated this week
- Your GeForce runs AI at half speed. This fixes it for 20/30/40 series cards. Up to 2× FP16 / BF16 / TF32 tensor throughput☆18,505Oct 29, 2026Updated 2 weeks ago
- Run GLM-5.3 (744B), GLM-5.3-Flash (321B), DeepSeek-V4.1-Flash (748B) and MiniMax-M3.1-Flash (428B) on your own GPUs - one engine, one das…☆2,827Updated this week
- High-performance single-GPU inference for selected model checkpoints and GPUs.☆3,332Updated this week
- An optimized quantization and inference library for running LLMs locally on modern consumer-class GPUs☆1,812Updated this week
- What I run on my inference server: Qwen3.8-Flash-Next on two RTX 3090s, and how I got here☆18Oct 12, 2026Updated last month
- Fastest validated sglang recipe for Qwen3.8-27B (AWQ-INT4 + DSpark) on RTX 3090s — 1×/2×/4× GPU presets, image+video, full benchmarks☆64Oct 17, 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…☆43,454Updated this week
- End-to-end encrypted cloud storage - Proton Drive • AdSpecial offer: 40% Off Yearly / 80% Off First Month. Protect your most important files, photos, and documents from prying eyes.
- llama.cpp/ik_llama.cpp launcher: loads big MoE models across mismatched multi-GPU rigs by exact VRAM math.☆284Updated this week
- DeepSeek-V4.1-Flash-1017 (552B MoE + 196B Engram) on 2x RTX 3090 + 128 GB RAM. Rust + CUDA, T16 quantization. 40+ tok/s measured with exp…☆100Nov 4, 2026Updated last week
- vLLM Docker Container for the latest Qwen 27b☆53Sep 27, 2026Updated last month
- A harness optimized to smaller LLMs☆2,725Sep 18, 2026Updated last month
- Deployment kit: Qwen3.8-27B EXL3 3.5bpw target + DFlash2 EXL3 5.0bpw speculative draft — launcher, config, model cards☆232Sep 1, 2026Updated 2 months ago
- RTX 6000 Pro Wiki — Running Large LLMs (Qwen3.5-397B, Kimi-K2.5, GLM-5) on PCIe GPUs without NVLink☆1,201Updated this week
- Production Docker deployment for Qwen3.8-27B AutoRound W4A16 (Vishva007) on 2x RTX 3090 (NVLink or PCIe). OpenAI-compatible vLLM with MTP…☆19Aug 20, 2026Updated 2 months ago
- 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.☆230May 14, 2026Updated 5 months ago
- Open quantization tooling for TurboQuant-style low-bit LLM releases, stock GGUF deployment, and Apple Silicon runtime experiments.☆10May 5, 2026Updated 6 months ago
- End-to-end encrypted cloud storage - Proton Drive • AdSpecial offer: 40% Off Yearly / 80% Off First Month. Protect your most important files, photos, and documents from prying eyes.
- Run Qwen3.8-27B at 140 tok/s on a single RTX 3090, up to 2x llama.cpp. Fast local LLM inference server with OpenAI- and Anthropic-compati…☆78Updated this week
- Frontier MoE models on RTX PRO 6000 / RTX 5090 (sm_120): DeepSeek-V4.1-Flash on 4x RTX PRO 6000, as a GPU+CPU hybrid on 4x RTX 5090, and …☆574Updated 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…☆4,192Updated this week
- vLLM 0.33 fork for consumer GPUs: 3 x 262K sessions on 4x RTX 3090, Qwen 3.8 Flash-Next at 207 tok/s, up to 17 on bigger cards. Seven che…☆49Updated this week
- One llama.cpp flag unlocks +33-39% decode speed for Qwen3.8-27B on consumer GPUs. The MTP head already ships inside your GGUF. Recipe, pa…☆370Sep 26, 2026Updated last month
- LLM inference in C/C++☆2,448Updated this week
- GVS5H: Five Qwen3.8-27B Models Match Claude Fable 5 on LiveCodeBench Hard | Fable 5 Level Coding for a Fifth the Price - or on a Single…☆1,205Nov 5, 2026Updated last week
Are these results useful?