A tiny LLM runtime microcore for KV cache, token budget, and batch scheduling.

LLMSched is not a full inference engine. It is a scheduler simulator that captures the core resource control plane of LLM serving: who gets KV cache, how many tokens to allocate, when to batch, and when to reject.

Inference resource scheduling authority — whoever controls token allocation, KV cache leases, and batch merging holds the key to inference performance.

No real model inference. Pure simulator:

requests.jsonl
 ↓
token budget
 ↓
KV cache allocation
 ↓
batch scheduling
 ↓
mock decode step
 ↓
trace.jsonl + scheduling report

cargo run -- run --requests examples/requests.jsonl

• Token Budget — global and per-request token allocation with overflow protection
• KV Cache Lease — allocate/free/evict semantics for GPU memory pages
• Batch Scheduler — priority queue + deadline-aware continuous batching
• Telemetry — structured trace output (JSONL) for downstream training

llmsched-core/
├── crates/llmsched-core/src/ # Core: queue, budget, kv, scheduler, telemetry
├── crates/llmsched-cli/src/ # CLI entry point
├── ffi/ # KernelLab C ABI bindings
├── examples/requests.jsonl # Sample workload
└── reports/ # Output trace + report

KernelLab → (C ABI) → LLMSched → (trace.jsonl) → ApexTrain-Core
 ↑
 Apeinx-IR → (plan.json)

TBD