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 fallback.
☆58Aug 8, 2026Updated 3 months ago
Alternatives and similar repositories for minimax-h3-fp16-fix
Users that are interested in minimax-h3-fp16-fix are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- ComfyUI custom node: run MiniMax H3 at near-fp16 speed with near-fp32 numerical stability on GPUs without bf16/fp8 hardware (V100 sm_70)☆19Aug 22, 2026Updated 2 months ago
- ComfyUI MiniMax H3 inference optimizations for NVIDIA Tesla V100 (SM70)☆31Sep 29, 2026Updated last month
- A ComfyUI custom node enabling **Flash Attention 1** on legacy NVIDIA GPUs (Tesla V100, T4) that lack Compute Capability 8.0+ required by…☆33Feb 9, 2026Updated 8 months ago
- SGLang for cool dudes using V100☆83Updated this week
- vLLM fork for Tesla V100☆15May 31, 2026Updated 5 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.
- Fork of llama.cpp Nvida Volta V100 (tensor parallelism 4 x v100 GPU)☆12Updated this week
- High-performance single-GPU inference for selected model checkpoints and GPUs.☆261Sep 14, 2026Updated last month
- Implementation of FlashAttention-2 for Nvidia Tesla V100 / Titan V☆217Updated this week
- NVIDIA Tesla V100 显卡驱动自动化管理套件☆33Oct 9, 2025Updated last year
- High-performance Duo-V100-GPU inference for selected model checkpoints and GPUs.☆37Updated this week
- forked from vllm-project/flash-attention☆74May 9, 2026Updated 5 months ago
- 1CatV2 with TileLANG written FA-v100 and many goodies☆20Jul 15, 2026Updated 3 months ago
- GGUF Quantization support for native ComfyUI models☆106Oct 30, 2026Updated last week
- 这是一套 MiniMax H3 本地部署与性能优化整理：在 4 张 22 GB RTX 2080 Ti 上，通过 Raylight FSDP/USP、双组NVLink、多卡 Qwen 文本编码节点、SM75 SageAttention 和通信流水，让MiniMax H3 的…☆25Aug 12, 2026Updated 2 months ago
- GPUs on demand by Runpod - Special Offer Available • AdRun AI, ML, and HPC workloads on powerful cloud GPUs—without limits or wasted spend. Deploy GPUs in under a minute and pay by the second.
- Auto-built Docker image of 1Cat-vLLM for NVIDIA Tesla V100 (sm_70/Volta) with CUDA 12.8 — rebuilt daily when upstream releases☆12Jul 22, 2026Updated 3 months ago
- OpenReel Studio source snapshot☆34Aug 17, 2026Updated 2 months ago
- LLM inference in C/C++☆22Oct 25, 2026Updated 2 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…☆30Updated this week
- Language files for the Shopkeepers plugin☆14Oct 19, 2026Updated 3 weeks ago
- Hand-written NVFP4 W4A16 CUDA kernels for Volta☆181Aug 19, 2026Updated 2 months ago
- 个人工作台搭建skill☆43Aug 4, 2026Updated 3 months ago
- 学霸帝 Zero-to-CAD — AI驱动的CAD代码生成器，基于Qwen3-VL-2B微调模型☆12May 31, 2026Updated 5 months ago
- Strata port for Tesla V100 (sm_70): Qwen3.8-Flash-Next with 262K context and MTP☆131Updated 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.
- ☆13Jan 21, 2026Updated 9 months ago
- V100 / SM70-focused vLLM engineering fork for modern LLM inference.☆1,480Updated this week
- Fine, I'll just put my pull requests here.☆12Jan 3, 2023Updated 3 years ago
- Windows desktop screenplay writing app with Chinese-first UI, Hollywood formatting, FDX support, and local export tools.☆16Updated this week
- Custom nodes for ComfyUI - Text Permutation, Loop nodes, Parameter Overlay, Visual Widgets☆16Aug 15, 2026Updated 2 months ago
- Emotion-first AI actor performance prompt skill☆34Aug 5, 2026Updated 3 months ago
- ☆18Mar 18, 2024Updated 2 years ago
- NInfer on Tesla V100 (sm_70): Qwen3.8-27B deployment, build/convert scripts, systemd service, and cross-engine benchmark (vs llama.cpp). …☆53Sep 15, 2026Updated last month
- 本地优先的一站式 AI 视频工作台☆41Oct 14, 2026Updated 3 weeks ago
- Virtual machines for every use case on DigitalOcean • AdGet dependable uptime with 99.99% SLA, simple security tools, and predictable monthly pricing with DigitalOcean's virtual machines, called Droplets.
- Block-FP8 (W8A16) on NVIDIA V100 / Volta (sm_70) for vLLM — the FP8 path upstream drops on Volta, tuned for sparse MoE.☆15Jul 10, 2026Updated 3 months ago
- FlashAttention brought back to Tesla V100 — a deep llama.cpp fork: SM 7.0 D256 kernels, SplitKV3, q4_0 KV cache, DFlash2 speculative deco…☆19Nov 1, 2026Updated last week
- fecyo的小程序部分☆11Apr 28, 2021Updated 5 years ago
- ☆18Dec 2, 2024Updated last year
- Simple AI Gateway Skyrim Mod Gateway Server☆14Sep 12, 2023Updated 3 years ago
- （非官方）中国海洋大学本科毕业论文 Typst 模板☆16May 19, 2026Updated 5 months ago
- 我们基于 Reasonix 架构接入 DeepSeek API，完成了一体化的安装与应用方案。该方案避免了复杂的环境配置过程，用户下载并安装后即可直接使用，显著降低了上手门槛。优势：性能够用，价格便宜（感谢ds）☆15May 26, 2026Updated 5 months ago