| 12345678 |
- c1-prose decode 29.28 tok/s (min 27.68 max 29.51 sd 0.69) ttft 0.295s toks [512, 512, 512, 512, 512] distinct 4 err 0 2026-09-24T00:17:29.650519Z..2026-09-24T00:18:59.353194Z
- c1-code decode 44.65 tok/s (min 42.39 max 44.79 sd 0.90) ttft 0.299s toks [512, 512, 512, 512, 512] distinct 3 err 0 2026-09-24T00:18:59.353548Z..2026-09-24T00:19:58.752312Z
- c1-chat_sampled decode 28.87 tok/s (min 26.65 max 29.72 sd 1.13) ttft 0.276s toks [512, 512, 512, 512, 512] distinct 5 err 0 2026-09-24T00:19:58.752638Z..2026-09-24T00:21:29.856881Z
- c4 agg 63.78 tok/s (min 60.18 max 67.14) per-stream 19.46 2026-09-24T00:21:29.857129Z..2026-09-24T00:24:11.541728Z
- c1-prose accept_len 2.17 step 73.2 ms (n=31 intervals, bs=1)
- c1-code accept_len 3.27 step 74.0 ms (n=20 intervals, bs=1.0)
- c1-chat_sampled accept_len 2.23 step 77.8 ms (n=29 intervals, bs=1)
- c4 accept_len 2.50 step 133.8 ms (n=33 intervals, bs=4)
|
