Taleef7/turboquant

Evidence-first KV-cache compression for Qwen2.5: packed low-bit KV blocks, PyTorch references, Triton kernels, and reproducible GPU benchmarks.

Stars: 0
Forks: 0
Open beginner issues: 0
Indexed issues: 5

Dominant language: Python
License: MIT
Last GitHub push: Aug 2, 2026
Latest indexed: Sep 20, 2026
Contributing guide: Contributing guide
Code of conduct: No code of conduct
Beginner labels: No beginner labels indexed
PR merge metrics: No merged PRs in 30d

Beginner friendly Everything

Open All

5 open issues indexed

Documentation, reproducibility, and README polish Open

documentation

Difficulty 4/5 3-5 days Newbie friendliness 48/100

Taleef7/turboquant#5 · 2 comments · Opened Apr 1, 2026
Long-context retrieval validation (Qwen scope closed through 32K) Open

enhancement testing

Difficulty 5/5 Over a week Newbie friendliness 20/100

Taleef7/turboquant#1 · 1 comment · Opened Apr 1, 2026
Bit-packing optimization (Phase 3) Open

enhancement optimization

Difficulty 5/5 Over a week Newbie friendliness 35/100

Taleef7/turboquant#2 · Opened Apr 1, 2026
Layer-adaptive quantization (Phase 4) Open

enhancement optimization

Difficulty 5/5 Over a week Newbie friendliness 25/100

Taleef7/turboquant#3 · Opened Apr 1, 2026
CUDA kernel optimization for higher throughput Open

enhancement performance

Difficulty 5/5 Over a week Newbie friendliness 35/100

Taleef7/turboquant#4 · Opened Apr 1, 2026

A short digest of beginner-friendly GitHub issues.