Implemented batching support for footless-orcasaq-2-27B-nvidia-p100. Benchmarks show 40 tok/s at concurrency = 4 running OrcaSAQ-2-27B (@OrcaRouter) on a single NVIDIA Tesla P100. More updates to follow. • Repo: • Demo (pure prose):

Maye is an inspiration. During tough moments I think about the resilience she’s shown and realize she’s encountered 100x the difficultly I’ve faced and she made it through with grace! It gives me resolve. She’s raised 3 children who’ve given such wondrous gifts to the world, so have tried to learn from her mothering instincts as well. I would highly recommend this book to anyone who wants a beacon of resilience to draw inspiration from (or really anyone, it’s an amazing read!) Congrats on today and thank you for being such an incredible Ouma to your grandkids and inspiration!

Testing MiniMax-M3.1-Flash-Preview by @MiniMax_AI on kernel optimization. Promising results so far, comparable to DeepSeek V4.1 Flash, but in my experience it trades a little speed for slightly sharper outputs.

Alright, so I just got OrcaSAQ-2-27B (Qwen3.8-27B quantized by orcarouter) up and running on an Nvidia P100. Obviously, performance isn't the greatest, but I honestly didn't expect to run a modern-architecture model using trellis-based quantization on a 2016 GPU with no Tensor Cores that currently goes for around $100–$200 on the used market. Pretty incredible. Status: No MTP, fine-tuning required, PPL matches model documentation. - GPU: Nvidia P100 16GB - Model: OrcaSAQ-2-27B by @OrcaRouter - Engine: footless (In-house development, currently private) Kernels & runtime: Built with assistance from DeepSeek v4.1 Flash by @deepseek_ai .