TrackerMetricsprocessing-ms-max

Backend processing time processing-ms-max

Maximum of the `openai-processing-ms` header within the prompt window (`har.processing_ms_max`). Measures the server side, not network or rendering. Measured 22 Aug 2026: 92–180 ms per single request; the search itself runs inside the SSE stream and is not included.

57 data points since 20 Aug 2026 · Download: JSON · CSV

Median across prompts, per mode

09118220 Aug 2026: Baseline (extractor, prompt-set, account)instant 20 Aug 2026: 116instant 21 Aug 2026: 112instant 22 Aug 2026: 118thinking 20 Aug 2026: 129thinking 21 Aug 2026: 182thinking 22 Aug 2026: 10320 Aug21 Aug22 Aug
instant thinking

Values per prompt

Prompt20 Aug 202621 Aug 202622 Aug 2026
K1-01 instant2020112128
K1-01 thinking129182129
K1-02 instant156
K10-01 instant139
K11-01 instant107
K11-02 instant180
K12-01 instant137
K12-02 instant127
K13-01 instant145
K14-01 instant8792
K14-01 thinking225103
K14-02 instant8898
K14-02 thinking15998
K2-01 instant106124123
K2-01 thinking100180134
K2-02 instant94
K2-03 instant157
K3-01 instant118
K3-02 instant97
K4-01 instant474130137
K4-01 thinking95175103
K4-02 instant102
K5-01 instant94103104
K5-01 thinking150185102
K5-02 instant112
K6-01 instant116133104
K6-01 thinking151204108
K6-02 instant115
K7-01 instant101
K7-02 instant106
K8-01 instant126
K9-01 instant119
K9-02 instant103