Unlock P2P comms between consumer NVIDIA GPUs
☆702Oct 26, 2026Updated 2 weeks ago
Alternatives and similar repositories for open-gpu-kernel-modules
Users that are interested in open-gpu-kernel-modules are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- NVIDIA Linux open GPU with P2P support☆1,443Updated 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…☆136Oct 15, 2026Updated 3 weeks ago
- ☆11Oct 18, 2026Updated 3 weeks ago
- Community recipes for serving LLMs on RTX 3090/4090/5090 CUDA gpus. Multi-engine (vLLM, sglang, llama.cpp and other custom engines) and m…☆2,482Updated this week
- An optimized quantization and inference library for running LLMs locally on modern consumer-class GPUs☆1,806Updated 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.
- Source-available C++/CUDA inference engine for Qwen3.8-27B INT4 on 3x RTX 3090☆44Updated this week
- Enable true multi gpu capability in Comfy UI using XDiT XFuser and FSDP managed by Ray☆477Updated this week
- Qwen3.8-Flash-Next (MoE) and Qwen3.8-27B on one or two 16 GB RTX 50-series GPUs: experts offloaded to CPU, two-GPU tensor parallel. 64 GB…☆140Updated this week
- Optimized VLLM for Ampere☆25Oct 18, 2026Updated 3 weeks ago
- 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 …☆2,631Updated 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…☆48Updated this week
- llama.cpp fork with additional SOTA quants and improved performance☆3,379Updated 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…☆507Oct 31, 2026Updated last week
- Lsglang is a special extension of sglang that fully utilizes CPU and GPU computing resources with an efficient GPU parallel + NUMA parall…☆168Updated this week
- Managed Database hosting by DigitalOcean • AdPostgreSQL, MySQL, MongoDB, Kafka, Valkey, and OpenSearch available. Automatically scale up storage and focus on building your apps.
- A tool to unlobotomize your NVIDIA card!☆836Oct 28, 2026Updated last 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☆18,076Oct 29, 2026Updated last week
- Practical local LLM recipes and benchmarks for RTX 5060 Ti setups☆202Oct 8, 2026Updated last month
- KVarN, KV cache precision tail, low-bit quants in llama.cpp for longer context of better precision in the same VRAM☆1,188Updated this week
- Xolotl: run DeepSeek-V4.1-Flash, all 552B, as a local coding agent on one gaming PC (RTX 4090 + 110 GB RAM): ~33 tok/s on agent tasks, ~4…☆495Updated this week
- LvLLM is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features …☆475Updated this week
- Qwen3.8 Flash Next AWQ INT4 4x3090 v1: 262K, Vision, FP8 QSA KV and CUDA Graphs on 4x RTX 3090☆16Aug 30, 2026Updated 2 months ago
- High-performance single-GPU inference for selected model checkpoints and GPUs.☆3,293Updated this week
- Multi-GPU prefill acceleration for llama.cpp☆16Oct 23, 2026Updated 2 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☆57Apr 28, 2026Updated 6 months ago
- Local LLM Inference Speed Test Tool☆250Oct 21, 2026Updated 2 weeks ago
- 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
- V100 / SM70-focused vLLM engineering fork for modern LLM inference.☆1,487Updated this week
- Tuning and qualification harness for the NVIDIA CMP 170HX (GA100). Measures, gates and recovers clock/voltage settings, and treats a comp…☆96Sep 28, 2026Updated last month
- Hand-written NVFP4 W4A16 CUDA kernels for Volta☆181Aug 19, 2026Updated 2 months ago
- LLM speculative inference server for heterogeneous hardware & consumer GPUs☆2,981Updated 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…☆96Updated this week
- Linux P2P unlock and performance fixes for NVIDIA CMP 170HX☆39Sep 22, 2026Updated last month
- 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.
- Make FP4 on 5090 Great Again☆24Jul 20, 2026Updated 3 months ago
- Run GLM-5.3-Flash (321B), DeepSeek-V4.1-Flash (748B) and MiniMax-M3.1-Flash (428B) on your own GPUs - one engine, one dashboard, OpenAI- …☆2,727Updated this week
- Dogfooding vLLM backport for older GPUs.☆359Updated this week
- llama.cpp fork for significantly improved performance on Ampere generation Nvidia cards (especially RTX 3090 / 3090 Ti): example: 110+ to…☆302Updated this week
- The official API server for Exllama. OAI compatible, lightweight, and fast.☆1,516Updated this week
- Multi-GPU (tensor-parallel) CUDA megakernel for Qwen3.8-27B GGUF (Q4_K_M, Q5_K_M, Q6_K, Q8_0) with on-device MTP speculative decoding. Fo…☆16Oct 23, 2026Updated 2 weeks ago
- PCIe P2P on consumer RTX 5090 via one GSP firmware registry key — no driver patches.☆25Apr 17, 2026Updated 6 months ago
Are these results useful?