Running LLM Tokenizer Arena & Context Inspector ⚡ Benchmark tokenizers & KV-cache across modern open LLMs