surya-koritala/Glyd
Lossless AI compression: 33% less GPU memory, bit for bit.
Updated 10 min ago
Growth
- Last 24 hours
- +33
- stars
- Last 7 days
- +86
- stars
- Last 30 days
- +324
- stars
- Total stars
- 334
- Forks
- 2
Top 3 in CLI tools this week
- 1 morluto/rea +116k stars this week
- 2 xerj-org/xerj +8.3k stars this week
- 3 kurelas/text-humanizer +721 stars this week
This repo: #46 of 265 this week, #11 of 51 today, #25 of 579 this month.
Also this week: #94 in LLM, #182 in Rust
See all 265 CLI tools reposStar history
Total stars at the end of each week, from GitHub’s daily star counts.
Why it’s rising: +86 stars this week, against ~9 in a usual week
Health signals
Too new to assessToo new to assess: the repository is less than 60 days old or has fewer than 10 commits. The signals below are shown for information only.
- Commit activity
- Commits every week since it was created (5 weeks) 1,172 commits on the default branch in 90 days
- Maintainers
- 1 commit author in 90 days Top author: 100% of the last 100 commits
- Issue responsiveness
- Too few issues to tell 0 issues opened by others in 90 days
- Pull requests
- 48 merged, 2 closed, 0 open Of the latest 50 opened in 90 days
- Releases
- Last release: 3 days ago 50+ releases in the last 12 months
- Project hygiene
- License, CI and security policy
- OpenSSF Scorecard
- Not scanned Unknown, not counted against the project
Signals are computed automatically from public GitHub data. They are not an endorsement or a security audit. How health is computed
README
Glyd
Lossless AI compression: 33% less GPU memory, bit for bit.
Glyd keeps a bf16 model's weights and KV cache in fewer bits in GPU memory and rebuilds the exact values as the model runs. It is the same model in about a third less memory, so it fits a smaller GPU or serves more requests on the same one.