Knowledge

• computer science 50

  • ai llm engineering 29
  • api design 2
  • data engineering 1
  • database engineering 2
  • design patterns 1
  • devops 3
  • distributed systems 2
  • frontend 1
  • hardware 5
  • os cs fundamentals 1
  • protocols 1
  • quantum computing 1
  • web security 1

Recent

• 2026-10-24 Concurrency (max_num_seqs): how many requests decode at once
• 2026-10-24 Context length (max_model_len): the per-request token cap
• 2026-10-24 Drafters: draft models, Medusa, EAGLE, MTP, DFlash, DSpark
• 2026-10-24 KV offload: a RAM or disk tier behind the KV pool
• 2026-10-24 KV pool: kv_cache_memory_bytes vs gpu_memory_utilization
• 2026-10-24 KV type: kv_cache_dtype
• 2026-10-24 Memory budget: what one start must fit
• 2026-10-24 vLLM settings, explained
• 2026-10-24 Prefill budget: tokens per engine step
• 2026-10-24 Prefix caching (--enable-prefix-caching)