1115714829 / llama.cpp-v100

llama.cpp optimized for NVIDIA Tesla V100 (SM70): Qwen3.8-Flash-Next with MTP (prefill 3402 tok/s at 200K context on 6x V100-16GB; also runs on 2x V100-16GB) and Qwen3.8-27B with DFlash2 (decode up to 220 tok/s); 2-6 GPUs, 262K-512K context, multimodal, concurrent serving, built-in web UI.

poisonxa16 / pxa

The set-and-forget LLM engine for Pascal and Volta: PXQ codec + kernels, auto-tuned per card. Ready-to-run PXQ models in MODELS.md; benchmarks vs llama.cpp in the README.