STACKQUADRANT

feifeibear/long-context-attention

Fine-tuning Tools

USP: Unified (a.k.a. Hybrid, 2D) Sequence Parallel Attention for Long Context Transformers Model Training and Inference

5.7
GitHub Metrics
Stars
681
Forks
81
Open Issues
13
Watchers
5
Contributors
24
Weekly Commits
0
Language
Python
License
Apache-2.0
Last Commit
May 21, 2026
Created
Mar 27, 2024
Latest Release
0.6.4
Release Date
Jan 13, 2026
Synced: Jul 18, 2026
Quality Scores
Documentation Qualityw: 20%
4.7

No dedicated docs site. Description: 119 chars. Stars signal: 681. Contributors: 24. Score: 4.7/10

Community Healthw: 20%
5.0

Stars: 681. Contributors: 24. Watchers: 5. Forks: 81. Issue ratio: 1.9%. Score: 5/10

Maintenance Velocityw: 15%
5.9

Last commit: 58d ago. Weekly commits: 0. Latest release: 0.6.4. Maturity bonus: 2.3y old. Score: 5.9/10

API Design & DXw: 20%
6.7

Stars/issues ratio: 52. Dynamic language: Python. No dedicated API docs. Permissive license: Apache-2.0. Popularity signal: 681 stars. Score: 6.7/10

Production Readinessw: 15%
5.9

Battle-tested: 681 stars. Peer review: 24 contributors. Versioned: 0.6.4. Licensed: Apache-2.0. Age: 2.3 years. Maintenance: last commit 58d ago. Score: 5.9/10

Ecosystem Integrationw: 10%
6.5

Fork interest: 81. Major ecosystem: Python. Integration-friendly: Apache-2.0. Adoption: 681 stars. Score: 6.5/10

Tags
attention-is-all-you-needdeepspeed-ulyssesllm-inferencellm-trainingpytorchring-attention
Radar
Documentation Quality
Community Health
Maintenance Velocity
API Design & DX
Production Readiness
Ecosystem Integration