AnswerDotAI/cold-compress
View on GitHubCold Compress is a hackable, lightweight, and open-source toolkit for creating and benchmarking cache compression methods built on top of GPT-Fast, a simple, PyTorch-native generation codebase.
- Stars
- 153
- Forks
- 16
- Open beginner issues
- 0
- Indexed issues
- 10
- Dominant language
- Python
- License
- BSD-3-Clause
- Last GitHub push
- Aug 9, 2024
- Latest indexed
- Sep 14, 2026
- Contributing guide
- No contributing guide
- Code of conduct
- No code of conduct
- Beginner labels
- No beginner labels indexed
- PR merge metrics
- No merged PRs in 30d
-
AnswerDotAI/cold-compress#29 · 0 comments · 0 reactions · 2 assignees ·
-
Implement InfLLM Open
AnswerDotAI/cold-compress#30 · 0 comments · 0 reactions · 1 assignee ·
-
AnswerDotAI/cold-compress#35 · 0 comments · 0 reactions · 0 assignees ·
-
Implement ThinK Open
AnswerDotAI/cold-compress#36 · 0 comments · 0 reactions · 0 assignees ·
-
AnswerDotAI/cold-compress#41 · 1 comment · 0 reactions · 0 assignees ·
-
SnapKV Open
AnswerDotAI/cold-compress#42 · 2 comments · 0 reactions · 0 assignees ·
-
AnswerDotAI/cold-compress#43 · 0 comments · 0 reactions · 0 assignees ·
-
AnswerDotAI/cold-compress#44 · 1 comment · 0 reactions · 0 assignees ·
-
Installation Issue Open
AnswerDotAI/cold-compress#45 · 0 comments · 0 reactions · 0 assignees ·
-
AnswerDotAI/cold-compress#46 · 0 comments · 0 reactions · 0 assignees ·