sigilant-runner · Phi-3.5-mini-instruct · A10G · vllm · 3 configs

Config | TPS | TPS p95 | TTFT | TTFT p95 | ITL | PPL | TPS% | TTFT% | PPL% | Score
INT8_W8A8 · ctx:16384 · kv:k16v16 · long  <- best | 52.9 | n/a | 906.5 | n/a | 18.96 | 2.88 | 100.0 | 100.0 | 100.0 | 100
INT8_W8A8 · ctx:16384 · kv:k8v8 · default | 52.9 | n/a | 1612.8 | n/a | 18.96 | 2.88 | 100.0 | 56.2 | 100.0 | 91
INT8_W8A8 · ctx:8192 · kv:k16v16 · default | 52.9 | n/a | 1637.2 | n/a | 18.97 | 2.88 | 100.0 | 55.4 | 100.0 | 91

Best config: INT8_W8A8 · ctx:16384 · kv:k16v16 · long
Auto baseline compare: score Δ=0.00 TPS Δ=0.00 TTFT Δ=0.0ms PPL Δ=0.00
Confidence: target=medium gap_before=9.00% var_before=n/a% replay=False(disabled) gap_after=9.00%

PPL is a quality proxy, not production validation.
Full production safety and long-context certification require Sigilant Optimizer.
