[ab] server http://172.17.0.1:8100 model /models/Qwen3.8-27B-exl3-4.0bpw
[transformers] Token indices sequence length is longer than the specified maximum sequence length for this model (317092 > 262144). Running this sequence through the model will result in indexing errors
  [0] code       prompt=8244 out=512 proposed=1045 accepted=303 rate=0.2900 tok/s= 59.97 length
  [1] code       prompt=8243 out=512 proposed=960 accepted=319 rate=0.3323 tok/s= 64.19 length
  [2] code       prompt=8243 out=512 proposed=840 accepted=346 rate=0.4119 tok/s= 71.12 length
  [3] code       prompt=8244 out=512 proposed=775 accepted=356 rate=0.4594 tok/s= 75.68 length
  [4] wiki       prompt=8244 out=512 proposed=875 accepted=340 rate=0.3886 tok/s= 69.13 length
  [5] wiki       prompt=8244 out=512 proposed=835 accepted=345 rate=0.4132 tok/s= 71.69 length
  [6] wiki       prompt=8244 out=512 proposed=815 accepted=349 rate=0.4282 tok/s= 72.86 length
  [7] wiki       prompt=8244 out=512 proposed=925 accepted=331 rate=0.3578 tok/s= 66.05 length
  [8] math       prompt=8244 out=512 proposed=685 accepted=376 rate=0.5489 tok/s= 82.97 length
  [9] math       prompt=8244 out=512 proposed=695 accepted=373 rate=0.5367 tok/s= 82.08 length
  [10] math       prompt=8244 out=512 proposed=975 accepted=321 rate=0.3292 tok/s= 63.43 length
  [11] math       prompt=8244 out=498 proposed=775 accepted=346 rate=0.4465 tok/s= 73.61 stop
  [12] longctx    prompt=8244 out=512 proposed=995 accepted=314 rate=0.3156 tok/s= 62.29 length
  [13] longctx    prompt=8244 out=512 proposed=805 accepted=351 rate=0.4360 tok/s= 73.56 length
  [14] longctx    prompt=8244 out=512 proposed=990 accepted=314 rate=0.3172 tok/s= 62.63 length
  [15] longctx    prompt=8244 out=512 proposed=880 accepted=336 rate=0.3818 tok/s= 68.65 length
  [16] sci        prompt=8244 out=222 proposed=360 accepted=150 rate=0.4167 tok/s= 54.94 stop
  [17] sci        prompt=8244 out=512 proposed=650 accepted=381 rate=0.5862 tok/s= 86.15 length
  [18] sci        prompt=8244 out=333 proposed=525 accepted=228 rate=0.4343 tok/s= 65.02 stop
  [19] sci        prompt=8245 out=488 proposed=675 accepted=353 rate=0.5230 tok/s= 79.89 stop
[ab] pooled accept rate 0.4062 over 20 prompts -> /cache/ab/Gg.json

## pins

## pins (from the server log)
