perkel666 /
MegaCapybara
The fastest inference engine for Qwen3.8-27B on the NVIDIA RTX 5090: up to 500 tokens/s for one agent and up to 2,000 tokens/s for many, contexts up to 1M tokens, and a launcher that shows what every setting costs. Windows and Linux.
Updated 8 min ago
Why now
Version v1.38 came out on October 26.
+6 stars gained this week
Stars gained each day, last 30 days
Oct 4Oct 19Today
This week Before
In short
MegaCapybara includes a launcher that shows how each setting trades accuracy, speed and memory.
Summary written by AI from the repository’s description and README.
perkel666 / MegaCapybara
View on GitHub(opens in new tab)