A 21M-parameter LM with a 16.8M-row product-key memory: as good as a 114M dense model and runs with the table on an SSD. Triton kernels for ROCm and CUDA.
-
Updated
Oct 23, 2026 - Python
#
memory-layers
Here are 3 public repositories matching this topic...
A high-power, local-first Obsidian plugin that unifies isolated AI coding agent memories (Antigravity, Claude Code, OpenCode, Codex) into an interlinked spatial graph, cross-platform matrix, and prompt exporter.
opencode local-first obsedian memory-layer codex-plugin codex-skills antigravity-ide memory-layer-for-ai opencode-skills claudcode memory-layers aicodingagent obsedianplugin
-
Updated
Aug 6, 2026 - JavaScript
Independent reconstruction of Memory Layers at Scale (Berges et al., 2024) on a single 24GB consumer GPU — Qwen2.5-7B, frozen backbone, with honest lessons learned.
-
Updated
Aug 5, 2026 - Python
Add this topic to your repo
To associate your repository with the memory-layers topic, visit your repo's landing page and select "manage topics."