Database Mart vLLM benchmark, RTX A6000, 50 concurrent

Database Mart hosted-GPU report, page footer dated 10/07/2026. Engine is vLLM built from github.com/vllm-project/vllm with no release tag or commit printed, so the engine version is unpinned. Weights are the Hugging Face checkpoint at the precision the table labels 'Quantization 16' (bf16/fp16 not distinguished); format_id is null because data/quantization/ has no fp16/bf16 record. Single GPU, tensor parallel 1. Input length fixed at 100 tokens, output length fixed at 600 tokens. The 'Output (tokens/s)' column is aggregate generated tokens across the concurrent requests divided by benchmark duration, NOT a per-request decode rate. Median TTFT is the median time to first token under that same concurrency. Third-party measurement by the GPU host; confidence capped because the harness and the vLLM commit are unpublished. Card: the page says 'Nvidia Quadro RTX A6000' but the spec block (Ampere, compute capability 8.6, 10,752 CUDA cores, 336 tensor cores, 48 GB GDDR6) is the RTX A6000, not the Turing Quadro RTX 6000, so records attach to nvidia-rtx-a6000. ONLY the 50-concurrent scenario is recorded: the '100 concurrent' table prints Request Numbers = 50 while Total Input Tokens = 10000, so its concurrency is self-contradictory and was dropped.