TinyEngram
Research of DeepSeek Engram Architecture based on Qwen-3 and Stable Diffusion series
GraphCanon updated 1d · GitHub synced 1d
Decision brief
TinyEngram is dedicated to researching the DeepSeek Engram architecture using Qwen-3 and Stable Diffusion for fine-tuning and memory injection tasks related to LLMs.
Good fit when
- - When you are specifically exploring or working on projects involving the DeepSeek Engram architecture
- - For tasks such as fine-tuning a model where precise control over model memory injection is required using Qwen-3 or Stable Diffusion components
Avoid when
- - If your project does not require the unique capabilities of the DeepSeek Engram architecture, as TinyEngram focuses exclusively on this framework
- - When only general-purpose LLM training and fine-tuning are needed without the specialized features provided by Qwen-3 or Stable Diffusion
Observed Jul 16, 2026 · Source: enrich:decision_facts
Verify the decision
Maintenance and security
Full trust report- Maintenance
- Slowing (95d since push)
- As of 1d
- Provenance
- Not a fork · Organization account
- As of 1d
- Security (OSV)
- No lockfile
- As of 1mo
Public GitHub metadata and optional OSV scans. Signals, not a guarantee. Trust methodology.
Install
pip install TinyEngramSimilar tools
Same-category neighbours. No typed graph edges are catalogued for this tool yet.
Evidence and technical details
Sourced facts, taxonomy, compatibility claims, README excerpt, and machine-readable endpoints.
Overview
TinyEngram repository focuses on research related to the DeepSeek Engram architecture using components like Qwen-3 and Stable Diffusion for tasks such as fine-tuning, model memory injection, and other LLM-related operations.
Capability facts
- Languages
- python
Source: github.language · Aug 24, 2026
Categories
Compatibility
Sourced claims from the README excerpt - not unsourced marketing copy.
Tags
README
TinyEngram: Exploring New Axis of Scaling and Memory Injection
Open research on DeepSeek-AI's Engram and memory injection in Qwen, StableDiffusion and more.
[!NOTE] TL;DR: TinyEngram demonstrates that Engram-based memory injection outperforms LoRA in both parameter efficiency and catastrophic forgetting resistance—and extends seamlessly to vision (e.g., Stable Diffusion) for lightweight, composable concept injection. All code, logs, and experiments are open!
If you find TinyEngram useful, a ⭐ helps support the project.
📢 Latest Announcements
- 2026.05.20 — 📝 TinyEngram-Vision technical report is ready. We organized the vision findings into a complete technical report to invite discussion and further exploration. Read the report, or check out the arXiv version if you prefer.
- 2026.02.12 — 🖼️ TinyEngram meets Vision! We injected visual concepts into Stable Diffusion through Engram, check our new cross-modal experiment!
- 2026.02.02 — 📌 Released reproduction scripts for Engram vs LoRA experiment.
- 2026.01.30 — 📌 Added comparison of catastrophic forgetting between TinyEngram and LoRA.
- 2026.01.30 — 📌 Added parameter ablation studies of TinyEngram with convergence observations.
- 2026.01.23 — 🎉 Initial TinyEngram commit.
🔍 Quick Navigation
Key Finding 1: Engram as Parameter Efficient Fine-Tuning Method
For agents
This page has a .md twin and JSON over the API.