Base44 Superagents-AI agent that does it all
Mini toolsBrumby-14B-Base
By Manifest AI
Brumby-14B-Base keeps a transformer-like block structure but swaps attention for a recurrent Power Retention mechanism that stores and updates information over arbitrarily long sequences without quadratic memory growth. This attention-free 14B model matches or approaches state-of-the-art open transformers at similar scale on many benchmarks while being more hardware friendly, and serves as the first in a planned Brumby family from 1B to 100B parameters.
New
Text
Gen 7
Overview
Brumby-14B-Base is Manifest AI’s 14B attention-free language model, a retrained Qwen3-14B variant that replaces attention with Power Retention layers for hardware-efficient, long-context reasoning under Apache 2.0.
Tools using Brumby-14B-Base
No tools found for this model yet.
Last updated: February 25, 2026
