github.devotd.

ohgeebtw/geezuz-qwen3.8-27b-dual-gpu

A repository that runs Qwen3.8-27B GGUF on two GPUs with a tensor-parallel CUDA megakernel, and supports on-device speculative decoding with a shared KV cache and prefix cache.

View on GitHub ↗
stars
★ 7
Forks
2
Language
Cuda
License
AGPL-3.0
Latest release
v0.3.2

Posts

1