Cross-tier CPU/GPU hybrid MoE inference for SGLang (plugin, tracks upstream main)
☆83Nov 19, 2026Updated this week
Alternatives and similar repositories for xtier
Users that are interested in xtier are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- XTU = X Transformers Unity — hybrid (CPU-expert + GPU) inference acceleration for vLLM: mainline plugin + CPU MoE engine (AVX512-VNNI/BF1…☆16Nov 2, 2026Updated 2 weeks ago
- 1Cat Studio: unified vLLM workbench, Codex Agent and creative canvas☆15Nov 6, 2026Updated 2 weeks ago
- A fork of jasl/vllm (codex/ds4-sm120-min-enable) with CPU-GPU hybrid inference support for DeepSeek-V4 on SM120+.☆32Sep 10, 2026Updated 2 months ago
- Max-context, high-speed Qwen3.8-27B on a single 5090: SGLang + NVFP4 + drafter, with monitoring.☆14Sep 8, 2026Updated 2 months ago
- ☆36Oct 20, 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.
- LvLLM is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features …☆476Nov 6, 2026Updated 2 weeks ago
- Développement World of Draenor Custom sur trinity 6x .☆17Updated this week
- ☆115Updated this week
- NVIDIA Linux open GPU with P2P support☆48Oct 26, 2026Updated 3 weeks ago
- ☆20Oct 29, 2026Updated 3 weeks ago
- Guide to using an NVIDIA OAM adapter card to connect an Intel Gaudi2 to PCIe and get it properly driven and working under Ubuntu☆31Nov 4, 2026Updated 2 weeks ago
- Run AMD AITER's hand-written ASM kernels on CDNA2 / gfx90a (MI210, MI250). 242 of 1,422 kernels translated, with the tests and benchmarks…☆11Sep 9, 2026Updated 2 months ago
- KVMem virtualizes KV memory across inference backends to make ultra-long-context local AI accessible on limited hardware.☆1,360Updated this week
- RTX 5070 Ti Laptop 12GB 跑 Qwen3.8-27B:NInfer + KVMem、256K 逻辑上下文、RAM KV、MTP、快速缓存续接与 dsh 本地 Agent / 搜索 / 浏览器。含源码、Windows 部署及实测。☆104Updated 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.
- 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…☆542Updated this week
- 一个几乎由 AI 编码的AI 便签工具☆21Jan 2, 2026Updated 10 months ago
- C++/CUDA single-GPU inference engine for Qwen3.8 (incl. Qwen3.8-Flash-Next with offloaded experts) on the RTX 5090; grew from NInfer☆308Updated this week
- 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…☆3,061Updated this week
- Strata for a PC with two NVIDIA GPUs and 32 GB of RAM: the low-RAM mode on a layer split, both cards working at the same time. 143 tok/s …☆85Oct 6, 2026Updated last month
- dlmrt: a single-request inference engine for block-diffusion language models☆18Nov 3, 2026Updated 2 weeks ago
- A reproducible local LLM benchmarking platform with versioned datasets, deterministic scoring, and auditable reports.☆183Oct 28, 2026Updated 3 weeks ago
- Dogfooding vLLM backport for older GPUs.☆14Oct 29, 2026Updated 3 weeks ago
- Reproducible Docker/vLLM deployment, tuning, and benchmarks for DeepSeek V4.1 Flash on a switchless three-node or four-node DGX Spark fab…☆235Updated this week
- 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.
- Windows本地bonsai模型部署启动器☆38Oct 20, 2026Updated last month
- Live subtitles with translation using stream API for macOS☆14Aug 3, 2026Updated 3 months ago
- Single-card Intel Gaudi2 vLLM source snapshot: FP8 Qwen, non-greedy DFlash2/MTP, GDN kernels and host-side optimizations.☆19Updated this week
- dsv41-flash-pp5-170hx &pp6 Supports the original (full/unquantized) model. Prefill 7,000 t/s, decode 45-75-125 t/s. dspark. Requires 200 …☆13Updated this week
- some experiments about 170hx 10g version.☆14Oct 19, 2026Updated last month
- High-performance single-GPU inference for selected model checkpoints and GPUs.☆3,387Updated this week
- Unlock P2P comms between consumer NVIDIA GPUs☆735Oct 26, 2026Updated 3 weeks ago
- ☆598Oct 20, 2026Updated last month
- A plugin for Obsidian that uses AI to help you write better and faster.☆12Apr 23, 2024Updated 2 years ago
- Managed Database hosting by DigitalOcean • AdPostgreSQL, MySQL, MongoDB, Kafka, Valkey, and OpenSearch available. Automatically scale up storage and focus on building your apps.
- Run MoE models beyond VRAM on AMD, NVIDIA & Intel GPUs. Vulkan-powered, portable Windows app with automatic setup. | 让游戏显卡跑起超显存大模型,支持 AMD…☆372Updated this week
- openwrt ndpi-netfilter2 packet☆10Apr 2, 2019Updated 7 years ago
- DeepSeek-V4-Flash-0731 tuned on 4x DGX Spark (GB10) at TP=4 — 123.13 tok/s decode. Every variable measured one at a time, with the values…☆13Aug 11, 2026Updated 3 months ago
- 1Cat-vLLM on Intel Gaudi☆48Oct 12, 2026Updated last month
- chinese-laya☆17Oct 13, 2026Updated last month
- ☆37Nov 2, 2026Updated 2 weeks ago
- Run DeepSeek-V4.1-Flash / DeepSeek-V4-Flash and GLM-5.3-Flash on SM89 (Ada / RTX 4090) and SM120 (RTX PRO 6000) with vLLM☆209Updated this week
Are these results useful?