volta

Dissecting NVIDIA GPU Architecture

Hand-written NVFP4 W4A16 CUDA kernels for Volta

RFCs for changes to Volta

Like Volta, but for Java — automatic per-project JDK switching via .java-version

PXQ: PXA-native low-bit MoE quants (2/3/4-bit, E16-row scales) + fused CUDA kernels for Pascal/Volta — run a real 35B on a salvaged 12-16GB card. Fork of ik_llama.cpp.

Benchmarks and notes for running modern LLMs with vLLM on 8x Tesla V100-32GB in 2026.

volta 简体中文版文档

Zsh plugin to seamlessly install and configure volta

Qwen3.8-27B in native NVFP4/FP8 on 2x PCIe Tesla V100-32GB (SM70): the PCIe runbook for v100-skinny + 1Cat-vLLM, with the 3 fixes that make it work without NVLink. 61-74 tok/s decode, MTP speculative decoding, OpenAI-compatible.

📦 A fully automated method for installing Nvidia drivers on Arch Linux

💀 Weekly base volta update to provide the prebuilt binary for aarch-linux-gnu openssl-1.1

Production repo for the volta.sh website.

A tui for asdf vm

FlashAttention brought back to Tesla V100 — a deep llama.cpp fork: SM 7.0 D256 kernels, SplitKV3, q4_0 KV cache, DFlash2 speculative decoding and multimodal fixes.

Multi-GPU acceleration for MiniMax H3 video generation on NVIDIA V100 (sm_70). Ulysses sequence parallelism as a drop-in ComfyUI custom node — ~19 min to ~7 min on 8x V100.

Run Qwen3.6-27B on four Tesla V100s at 366 tok/s using hand-written NVFP4 CUDA kernels and chain-MTP speculation.

⚡ A Docker image for Volta

Config for fast start zsh

ZSH Plugin to install and load Volta: JS Toolchains as Code. ⚡