Compare repositories
This page shows the Star History and the facts of 2 to 5 repositories in the Repository Index, side by side. You choose the repositories. The page does not rank them.
Facts
| Fact | BlinkDL/RWKV-LM RWKV (pronounced RwaKuv) is an RNN with great LLM performance, which can also be directly trained like a GPT transformer (parallelizable). We are at RWKV-7 "Goose". So it's combining the best of RNN and transformer - great performance, linear time, constant space (no kv-cache), fast training, infinite ctx_len, and free sentence embedding. | labmlai/annotated_deep_learning_paper_implementations 🧑🏫 60+ Implementations/tutorials of deep learning papers with side-by-side notes 📝; including transformers (original, xl, switch, feedback, vit, ...), optimizers (adam, adabelief, sophia, ...), gans(cyclegan, stylegan2, ...), 🎮 reinforcement learning (ppo, dqn), capsnet, distillation, ... 🧠 |
|---|---|---|
| Stars | 14,771 | 67,672 |
| Net stars, last 30 days | +42 | +206 |
| Largest star day | +8 on Jun 30, 2026 | +82 on May 12, 2026 |
| Milestone reached in the record | None | None |
| Forks | 1,026 | 6,781 |
| Language | Python | Python |
| License | APACHE-2.0 | MIT |
| Created | Aug 8, 2021 | Aug 25, 2020 |
| Last push | Oct 31, 2026 | Nov 2, 2026 |
| Archived | No | No |
| Health facts: Maintenance | 4 of 4 checks | 3 of 4 checks |
| Health facts: Releases | 0 of 3 checks | 0 of 3 checks (1 without data) |
| Health facts: Security practice | 0 of 1 check | 0 of 1 check |
| Health facts: Community files | 1 of 3 checks | 1 of 3 checks |
| Commits, last 13 weeks | 12 | 5 |
| Weeks with commits, last 52 | 32 of 52 | 4 of 52 |
| In the Repository Index since | Oct 22, 2026 | Oct 22, 2026 |
Topics that all of them have on the Repository Host: deep-learning, pytorch, transformer, transformers. The facts come from the Repository Host and from the Star History that AwesomeRepos records. A shared topic does not mean that one repository can replace another. Each Health Facts row is one dimension alone. AwesomeRepos does not add them into one score. How each check is defined