airllm
github.com/2233admin/airllm
aa2a6f6mainsynced 1mo ago
Code HealthView report
Average health
6.1/10
Fair
Hotspot health
10.0/10
Excellent
CostsView costs →
What this repo's indexing and doc generation cost, per sync.
Open the Costs page →Attention Needed20
air_llm/init.pyKnowledge Silo
100% single-owner concentration
air_llm/inference_example.pyDead Code
Safe to delete (0 lines)
air_llm/airllm/init.pyKnowledge Silo
100% single-owner concentration
anima_100k/modeling_flash_llama.pyDead Code
Safe to delete (440 lines)
air_llm/airllm/airllm.pyKnowledge Silo
100% single-owner concentration
No commit history
Commit activity appears once git history is indexed.
Recent DecisionsView all
Avoid reloading layers by caching intermediate activations in RAMproposed
· commentShard Llama model layers to reduce GPU memoryproposed
· commentMacOS support constrained to Apple Silicon and required dependenciesproposed
· readme miningProvide option to delete original downloaded model to save disk spaceproposed
· readme miningSupport CPU inference and non-sharded modelsproposed
· readme miningAuto-detect model type to reduce required model class specificationproposed
· readme miningNo hotspot data
No churn or complexity hotspots detected in this repository.