GitHub · Python
Coekjan/CrossPool
Efficient GPU Memory Pooling for Multi-LLM Serving via KV Cache and Weight Disaggregation
- Stars
- 11
- Forks
- 7
- Stars gained in 24 hours
- 0
- Forks gained in 24 hours
- +1
Topics: colocation, disaggregation, kv-cache, llm-inference, llm-serving, memory-disaggregation, memory-pooling
Latest data point: Nov 22, 2026 · 11 stars · 7 forks