v100

Code for testing the native float16 matrix multiplication performance on Tesla P100 and V100 GPU based on cublasHgemm

Hashcat Benchmark Comparison

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

vLLM for nVidia Volta (sm_70)

ComfyUI custom node: run MiniMax H3 at near-fp16 speed with near-fp32 numerical stability on GPUs without bf16/fp8 hardware (V100 sm_70)

Found out that using A100 and V100 on Vicuna and Llama2 have a different result, while other model such as Falcon doesn't has such question.

Performance of CUDA example benchmark code on NVIDIA A100.

SunFire V100 10 Inch Compact Case

Serve Qwen3.5-397B-A17B (AWQ) on 8x Tesla V100-SXM2-32GB (DGX-1, TP8) for agentic coding & ops — a downstream fork of 1Cat-vLLM.

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 Qwen3.6-35B-A3B NVFP4 on non-Blackwell GPUs via llama.cpp dequant (V100-tested)

Throughput-oriented Llama 3.1 405B inference on 2 Taiwania 2 nodes / 16 V100 GPUs: continuous batching, a V100-compatible FlashAttention backend, and cross-node tensor parallelism over NCCL InfiniBand/GDRDMA.

VastLLM: a production-oriented FastLLM fork for native C++ inference, V100/SM70, long context, and Qwen3.5/3.6; upstream: ztxz16/fastllm