[22:19:51Z] === codetwig-vllm-coldwarm-1 run_id=8d63125e-6074-4143-befd-1139add7a797 ===
[22:19:51Z] Target: http://127.0.0.1:18080 model=Qwen3.8-MXFP4 T=0.1 1 cold + 5 warm per prompt, sequential
[22:19:51Z] >>> CELL: r9700x2:vllm:qwen3.8-27b-mxfp4-dflash2-think (enable_thinking=True)
[22:19:51Z]   gpu snapshot: renderD128 33876262912 34208743424 17000000 210000000 | renderD129 33876434944 34208743424 16000000 210000000
[22:19:51Z]   prompt=hello cold
[22:19:52Z]     ttft=109ms total=268ms ct=39 tok/s=245.3 finish=stop accept=0.653 others=0.0 err=None
[22:19:52Z]   prompt=hello warm 1
[22:19:52Z]     ttft=100ms total=259ms ct=39 tok/s=245.3 finish=stop accept=0.653 others=0.0 err=None
[22:19:52Z]   prompt=hello warm 2
[22:19:53Z]     ttft=110ms total=264ms ct=39 tok/s=253.2 finish=stop accept=0.653 others=0.0 err=None
[22:19:53Z]   prompt=hello warm 3
[22:19:53Z]     ttft=104ms total=258ms ct=39 tok/s=253.2 finish=stop accept=0.653 others=0.0 err=None
[22:19:53Z]   prompt=hello warm 4
[22:19:54Z]     ttft=109ms total=263ms ct=39 tok/s=253.2 finish=stop accept=0.653 others=0.0 err=None
[22:19:54Z]   prompt=hello warm 5
[22:19:54Z]     ttft=99ms total=252ms ct=39 tok/s=254.9 finish=stop accept=0.653 others=0.0 err=None
[22:19:54Z]   prompt=P-MEDIUM cold
[22:19:58Z]     ttft=105ms total=3177ms ct=512 tok/s=166.7 finish=length accept=0.395 others=0.0 err=None
[22:19:58Z]   prompt=P-MEDIUM warm 1
[22:20:01Z]     ttft=98ms total=2769ms ct=480 tok/s=179.7 finish=stop accept=0.438 others=0.0 err=None
[22:20:01Z]   prompt=P-MEDIUM warm 2
[22:20:04Z]     ttft=98ms total=2517ms ct=459 tok/s=189.7 finish=stop accept=0.47 others=0.0 err=None
[22:20:04Z]   prompt=P-MEDIUM warm 3
[22:20:07Z]     ttft=98ms total=2654ms ct=434 tok/s=169.8 finish=stop accept=0.406 others=0.0 err=None
[22:20:07Z]   prompt=P-MEDIUM warm 4
[22:20:09Z]     ttft=118ms total=2441ms ct=428 tok/s=184.2 finish=stop accept=0.451 others=0.0 err=None
[22:20:09Z]   prompt=P-MEDIUM warm 5
[22:20:12Z]     ttft=102ms total=2776ms ct=463 tok/s=173.1 finish=stop accept=0.418 others=0.0 err=None
[22:20:12Z]   prompt=P-HARD cold
[22:20:18Z]     ttft=110ms total=4996ms ct=1024 tok/s=209.6 finish=length accept=0.543 others=0.0 err=None
[22:20:18Z]   prompt=P-HARD warm 1
[22:20:23Z]     ttft=102ms total=4747ms ct=1024 tok/s=220.5 finish=length accept=0.582 others=0.0 err=None
[22:20:23Z]   prompt=P-HARD warm 2
[22:20:28Z]     ttft=115ms total=5123ms ct=1024 tok/s=204.5 finish=length accept=0.525 others=0.0 err=None
[22:20:28Z]   prompt=P-HARD warm 3
[22:20:34Z]     ttft=102ms total=5415ms ct=1024 tok/s=192.7 finish=length accept=0.491 others=0.0 err=None
[22:20:34Z]   prompt=P-HARD warm 4
[22:20:39Z]     ttft=100ms total=4992ms ct=1024 tok/s=209.3 finish=length accept=0.545 others=0.0 err=None
[22:20:39Z]   prompt=P-HARD warm 5
[22:20:44Z]     ttft=108ms total=5384ms ct=1024 tok/s=194.1 finish=length accept=0.495 others=0.0 err=None
[22:20:45Z] >>> CELL: r9700x2:vllm:qwen3.8-27b-mxfp4-dflash2-nothink (enable_thinking=False)
[22:20:45Z]   gpu snapshot: renderD128 33876262912 34208743424 36000000 210000000 | renderD129 33876434944 34208743424 35000000 210000000
[22:20:45Z]   prompt=hello cold
[22:20:45Z]     ttft=97ms total=142ms ct=11 tok/s=244.4 finish=stop accept=0.643 others=0.0 err=None
[22:20:45Z]   prompt=hello warm 1
[22:20:45Z]     ttft=94ms total=136ms ct=11 tok/s=261.9 finish=stop accept=0.643 others=0.0 err=None
[22:20:45Z]   prompt=hello warm 2
[22:20:46Z]     ttft=95ms total=140ms ct=11 tok/s=244.4 finish=stop accept=0.643 others=0.0 err=None
[22:20:46Z]   prompt=hello warm 3
[22:20:46Z]     ttft=93ms total=141ms ct=11 tok/s=229.2 finish=stop accept=0.643 others=0.0 err=None
[22:20:46Z]   prompt=hello warm 4
[22:20:46Z]     ttft=102ms total=144ms ct=11 tok/s=261.9 finish=stop accept=0.643 others=0.0 err=None
[22:20:46Z]   prompt=hello warm 5
[22:20:47Z]     ttft=92ms total=138ms ct=11 tok/s=239.1 finish=stop accept=0.643 others=0.0 err=None
[22:20:47Z]   prompt=P-MEDIUM cold
[22:20:48Z]     ttft=101ms total=890ms ct=113 tok/s=143.2 finish=stop accept=0.318 others=0.0 err=None
[22:20:48Z]   prompt=P-MEDIUM warm 1
[22:20:49Z]     ttft=103ms total=1028ms ct=135 tok/s=145.9 finish=stop accept=0.328 others=0.0 err=None
[22:20:49Z]   prompt=P-MEDIUM warm 2
[22:20:50Z]     ttft=105ms total=1072ms ct=128 tok/s=132.4 finish=stop accept=0.282 others=0.0 err=None
[22:20:50Z]   prompt=P-MEDIUM warm 3
[22:20:51Z]     ttft=99ms total=795ms ct=111 tok/s=159.5 finish=stop accept=0.369 others=0.0 err=None
[22:20:51Z]   prompt=P-MEDIUM warm 4
[22:20:52Z]     ttft=97ms total=750ms ct=110 tok/s=168.5 finish=stop accept=0.399 others=0.0 err=None
[22:20:52Z]   prompt=P-MEDIUM warm 5
[22:20:53Z]     ttft=101ms total=912ms ct=123 tok/s=151.7 finish=stop accept=0.345 others=0.0 err=None
[22:20:53Z]   prompt=P-HARD cold
[22:20:57Z]     ttft=106ms total=3180ms ct=371 tok/s=120.7 finish=stop accept=0.25 others=0.0 err=None
[22:20:57Z]   prompt=P-HARD warm 1
[22:21:00Z]     ttft=111ms total=3521ms ct=370 tok/s=108.5 finish=stop accept=0.21 others=0.0 err=None
[22:21:00Z]   prompt=P-HARD warm 2
[22:21:04Z]     ttft=108ms total=3445ms ct=347 tok/s=104.0 finish=stop accept=0.193 others=0.0 err=None
[22:21:04Z]   prompt=P-HARD warm 3
[22:21:08Z]     ttft=103ms total=3270ms ct=348 tok/s=109.9 finish=stop accept=0.215 others=0.0 err=None
[22:21:08Z]   prompt=P-HARD warm 4
[22:21:11Z]     ttft=116ms total=3390ms ct=368 tok/s=112.4 finish=stop accept=0.222 others=0.0 err=None
[22:21:11Z]   prompt=P-HARD warm 5
[22:21:15Z]     ttft=101ms total=3630ms ct=363 tok/s=102.9 finish=stop accept=0.192 others=0.0 err=None
[22:21:15Z] === bench complete ===