This is a GGUF-formatted, NVFP4-quantized release of a 27-billion-parameter model (Qwopus 3.6 v2) that includes Multi-Token Prediction (MTP) capabilities. It is designed for efficient local execution via llama.cpp or compatible GGUF runtimes. The model targets users who need high-performance local LLMs on consumer hardware.
Qwopus3.6 27B V2 MTP is a Foundation models & chat project. It focuses on providing a highly optimized, quantized large language model for local CPU/GPU inference using llama.cpp. Qwopus3.6 27B V2 MTP is an open-source project aimed at AI developers. The project is open source (MIT). Qwopus3.6 27B V2 MTP is available on the command line.
Behind Qwopus3.6 27B V2 MTP is michaelw9999, and it first shipped in 2026. The project is developed in the open on GitHub with 38 stars and 84 commits in the last 90 days. It operates in a well-populated space: PulseGate tracks 5 similar projects. Key capabilities include GGUF Format, MTP, and NVFP4 Quantization.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do