vllm 6
- The Illusion of Determinism in Disaggregated Inference
- Two Schedulers, One SLO
- The KV Cache Has No ABI
- Crossing the Boundary: Custom Kernels and the C++/Python ABI in vLLM
- Building vLLM from Source: A Field Guide (with all the pitfalls)
- vLLM's op IR, or: where the inference engine meets the compiler