marlin_v100 是一个从 vLLM 主树中提取出来的最小 Marlin 独立开发工作区，聚焦于 Marlin dense 与 Marlin MoE 的源码开发、最小构建和轻量验证。它保留了核心 CUDA/C++ 实现、最小 Python 薄封装、生成器测试与主树回写映射，适合在不受主树全量构建干扰的情况下快速重构和验证 Marlin 相关改动。
☆26Jul 2, 2026Updated 3 months ago
Alternatives and similar repositories for marlin_v100
Users that are interested in marlin_v100 are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- forked from vllm-project/flash-attention☆69May 9, 2026Updated 5 months ago
- Sol-Attn (arXiv 2607.24027) sparse acceleration for MiniMax H3 on V100 — single-node ComfyUI plugin (embedded FP16Safe + keep-or-drop ker…☆33Aug 29, 2026Updated last month
- 1CatV2 with TileLANG written FA-v100 and many goodies☆20Jul 15, 2026Updated 3 months ago
- Mixed-precision acceleration patch for MiniMax H3 inference on NVIDIA V100☆39Sep 6, 2026Updated last month
- Run MiniMax H3 at native fp16 on pre-bf16 GPUs (V100/Volta) — exact-math fix for the black-frame overflows, ~11× faster than the fp32 fal…☆55Aug 8, 2026Updated 2 months 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.
- ComfyUI MiniMax H3 inference optimizations for NVIDIA Tesla V100 (SM70)☆28Sep 29, 2026Updated 3 weeks ago
- A ComfyUI custom node enabling **Flash Attention 1** on legacy NVIDIA GPUs (Tesla V100, T4) that lack Compute Capability 8.0+ required by…☆31Feb 9, 2026Updated 8 months ago
- Implementation of FlashAttention-2 for Nvidia Tesla V100 / Titan V☆214Oct 14, 2026Updated last week
- NInfer on Tesla V100 (sm_70): Qwen3.8-27B deployment, build/convert scripts, systemd service, and cross-engine benchmark (vs llama.cpp). …☆50Sep 15, 2026Updated last month
- SGLang for cool dudes using V100☆80Oct 5, 2026Updated 2 weeks ago
- Enables the ComfyUI MiniMax H3 reference-to-video workflow to use a stronger FL2VA model and validated Ref2VA tensor-overlay modes to unl…☆31Sep 25, 2026Updated 3 weeks ago
- xllama.cpp — Optimized LLM Inference for NVidia Tesla V100-SXM2-32GB☆17Apr 29, 2026Updated 5 months ago
- H3-aware latent upscaling for ComfyUI with automatic 32-pixel alignment.☆25Aug 17, 2026Updated 2 months ago
- Benchmarks and notes for running modern LLMs with vLLM on 8x Tesla V100-32GB in 2026.☆23Jul 10, 2026Updated 3 months ago
- 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.
- Hand-written NVFP4 W4A16 CUDA kernels for Volta☆175Aug 19, 2026Updated 2 months ago
- Optimize SGEMM step by step☆10Sep 4, 2024Updated 2 years ago
- Agentic-coding oriented inference engine for N x V100-SXM2, based on SGLang: GLM-5.3-Flash, Qwen3.8-Flash-Next, DeepSeek-V4.1-Flash, Mini…☆32Oct 7, 2026Updated 2 weeks ago
- INSPIRE: Intensity and Spatial Information-Based Deformable Image Registration☆12Jun 30, 2021Updated 5 years ago
- xy-自由画布☆32Jun 1, 2026Updated 4 months ago
- vllm混合推理扩展插件，支持多NUMA混合推理，单卡推理Qwen3-Next模型可达1000+ prefill☆34Nov 7, 2025Updated 11 months ago
- ☆12Dec 24, 2025Updated 9 months ago
- Tibis 是一款面向本地写作与 AI 工作流的 Markdown 桌面编辑器，基于 Electron + Vue 3 + TypeScript 构建。 它把文档编辑、本地文件管理和多模型配置放在同一个桌面应用里，适合用于写作、知识整理、技术文档草稿和 AI 辅助内容生产…☆24Sep 28, 2026Updated 3 weeks ago
- Lumivra 是一个干净、现代、富有设计感的 WordPress 主题。专为追求极简美 学和优雅排版的内容创作者、博客作者和企业网站打造。Lumivra is a clean, modern, and beautifully designed WordPress theme…☆14Mar 22, 2026Updated 7 months ago
- Deploy open-source AI quickly and easily - Special Bonus Offer • AdRunpod Hub is built for open source. One-click deployment and autoscaling endpoints without provisioning your own infrastructure.
- NVIDIA Tesla V100 显卡驱动 自动化管理套件☆31Oct 9, 2025Updated last year
- Slate — Local previs workbench for AI short films. One storyboard.json drives the whole pipeline: film analysis → script → shots → assets…☆48Oct 7, 2026Updated 2 weeks ago
- A browserless Agent Skill that creates polished visuals locally—no image-generation model required.☆22Oct 8, 2026Updated 2 weeks ago
- llama.cpp optimized for NVIDIA Tesla V100 (SM70): large MoE Qwen3.8-Flash-Next with MTP (prefill 3464 tok/s at 200K context, 6x V100-16GB…☆46Updated this week
- Modular prompt engineering tool for ComfyUI with wildcards, presets, auto-categorize, and LoRA Manager integration☆20Feb 5, 2026Updated 8 months ago
- 2023华为嵌入式软件大赛-算法组 冠军队伍题解☆13Mar 2, 2024Updated 2 years ago
- A C++ Implementation of "RIFT: Multi-Modal Image Matching Based on Radiation-Variation Insensitive Feature Transform".☆16Apr 15, 2022Updated 4 years ago
- High-performance single-GPU inference for selected model checkpoints and GPUs.☆242Sep 14, 2026Updated last month
- ☆15Jun 29, 2025Updated last year
- 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.
- Pancake Cooking Robot☆18Jul 13, 2026Updated 3 months ago
- AMD SEV-SNP Go unikernel example☆15Oct 9, 2026Updated last week
- YunChat是一款有趣且开源的聊天应用，基于 Springboot、Vite、TypeScript、Redis、Socket.IO、Kotlin等技术开发☆16Apr 9, 2026Updated 6 months ago
- See where local inference loses speed on Apple Silicon under Linux☆13Updated this week
- Self-hosted text-to-speech platform with multi-backend support, voice cloning, and a modern web UI.☆13Sep 25, 2026Updated 3 weeks ago
- Research MIL-to-HWX compiler for the M4 Apple Neural Engine☆16Sep 3, 2026Updated last month
- Windows code for communicating with a VC707 evaluation board running a MIAOW compute unit.☆11Oct 20, 2015Updated 11 years ago