forked from vllm-project/flash-attention
☆75May 9, 2026Updated 6 months ago
Alternatives and similar repositories for flash-attention-v100
Users that are interested in flash-attention-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:
- 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…☆27Jul 2, 2026Updated 4 months ago
- Implementation of FlashAttention-2 for Nvidia Tesla V100 / Titan V☆220Updated this week
- This project is specifically developed for V100, based on lmdeploy 0.12.1, and supports mainstream open-source models from Q4 2025 to Q1 …☆22Mar 18, 2026Updated 7 months ago
- A ComfyUI custom node enabling **Flash Attention 1** on legacy NVIDIA GPUs (Tesla V100, T4) that lack Compute Capability 8.0+ required by…☆34Feb 9, 2026Updated 9 months ago
- vLLM fork for Tesla V100☆15May 31, 2026Updated 5 months 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.
- Mixed-precision acceleration patch for MiniMax H3 inference on NVIDIA V100☆40Sep 6, 2026Updated 2 months ago
- Auto-built Docker image of 1Cat-vLLM for NVIDIA Tesla V100 (sm_70/Volta) with CUDA 12.8 — rebuilt daily when upstream releases☆13Jul 22, 2026Updated 3 months ago
- SGLang for cool dudes using V100☆86Updated this week
- 1Cat Studio: unified vLLM workbench, Codex Agent and creative canvas☆14Nov 6, 2026Updated last week
- ComfyUI MiniMax H3 inference optimizations for NVIDIA Tesla V100 (SM70)☆36Sep 29, 2026Updated last month
- 自托管多引擎语音与音频工作台:整合六种本地 TTS 引擎及 Edge-TTS 在线兜底,支持声音设计、克隆、多角色配音、语音转换与音频编辑。 | A self-hosted speech and audio workstation combining six local T…☆14Updated 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
- ComfyUI custom node: run MiniMax H3 at near-fp16 speed with near-fp32 numerical stability on GPUs without bf16/fp8 hardware (V100 sm_70)☆21Updated this week
- 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…☆60Aug 8, 2026Updated 3 months 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.
- LvLLM is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features …☆475Nov 6, 2026Updated last week
- High-performance single-GPU inference for selected model checkpoints and GPUs.☆272Sep 14, 2026Updated 2 months ago
- Sol-Attn (arXiv 2607.24027) sparse acceleration for MiniMax H3 on V100 — single-node ComfyUI plugin (embedded FP16Safe + keep-or-drop ker…☆34Aug 29, 2026Updated 2 months ago
- llama.cpp optimized for NVIDIA Tesla V100 (SM70). Qwen3.8-Flash-Next (MoE) with MTP on 1/2/3/6x V100-16GB at 262K context (6 GPUs: prefil…☆80Updated this week
- Run local LLMs on a Tesla V100-SXM2-16GB: prebuilt binaries, serve scripts, and setup for Windows and Linux (Qwen3 + Gemma 4, Claude Code…☆20Sep 1, 2026Updated 2 months ago
- A 198t/s and 353t/s qwen 3.8 27b inferencing engine for 2*v100 and 4*v100☆115Updated this week
- NInfer on Tesla V100 (sm_70): Qwen3.8-27B deployment, build/convert scripts, systemd service, and cross-engine benchmark (vs llama.cpp). …☆55Sep 15, 2026Updated 2 months ago
- DeepSeek-V4-Flash-0731 on 4x CMP 170HX (sm_80): 98 tok/s decode, ~5300 tok/s prefill. DSpark speculative decoding under pipeline parallel…☆119Aug 13, 2026Updated 3 months ago
- ☆25Jun 27, 2026Updated 4 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.
- DeepSeek-V4.1-Flash on 4× NVIDIA CMP 170HX with vLLM: up to 297 tok/s for one user, up to 868 tok/s at eight, 262K context, one-command c…☆33Nov 2, 2026Updated 2 weeks ago
- Strata port for Tesla V100 (sm_70): Qwen3.8-Flash-Next with 262K context and MTP☆150Updated this week
- vLLM for NVIDIA CMP 170HX (Ampere sm_80): fast, repeatable model serving on one card or several☆29Nov 2, 2026Updated last week
- ☆80Feb 19, 2024Updated 2 years ago
- ☆18Nov 8, 2026Updated last week
- gfx906 setup document☆14Sep 1, 2025Updated last year
- INT8-native vLLM fork☆29Updated this week
- GLM-5.3-Flash on 4× NVIDIA CMP 170HX with vLLM: up to 482.6 tok/s single-user, up to 943.7 tok/s at 8 users, 262K context, TP4 or PP4, on…☆91Oct 26, 2026Updated 3 weeks ago
- Flash Attention v2 for Pascal (P100, GTX 1080) and Volta (V100) GPUs — bringing O(N) memory attention to hardware the official flash-attn…☆31Nov 3, 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.
- NVIDIA Tesla V100 显卡驱动自动化管理套件☆34Oct 9, 2025Updated last year
- triton for AMD gfx906 GPUs, e.g. Radeon VII / MI50 / MI60☆48Dec 8, 2025Updated 11 months ago
- TurboQuant+ KV cache compression for vLLM. 3.8x smaller KV cache, same conversation quality. Fused CUDA kernels with automatic PyTorch fa…☆82Nov 6, 2026Updated last week
- ☆10May 17, 2025Updated last year
- Exact LLM inference engine written from scratch for AMD Strix Halo (gfx1151, ROCm/HIP). Qwen3.8-Flash-Next on one node or two replicas; D…☆30Updated this week
- The definitive vLLM runtime for dual RTX 2080 Ti 22GB + NVLink, delivering Qwen 27B local inference with maximum 200+ tok/s single-reques…☆1,222Oct 1, 2026Updated last month
- some experiments about 170hx 10g version.☆13Oct 19, 2026Updated 3 weeks ago
Are these results useful?