mudassiruddin21 commited on
Commit
b0c1dc8
·
verified ·
1 Parent(s): 0994ace

Run r3: logs and results

Browse files
runs/r3/engine_logs/20261119-043028-eval-r1.log CHANGED
@@ -300,3 +300,231 @@ INFO: 127.0.0.1:48033 - "POST /v1/completions HTTP/1.1" 200 OK
300
  INFO: 127.0.0.1:62756 - "POST /v1/completions HTTP/1.1" 200 OK
301
  (APIServer pid=43360) INFO 11-19 04:36:45 [loggers.py:320] Engine 000: Avg prompt throughput: 8.5 tokens/s, Avg generation throughput: 59.7 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
302
  (APIServer pid=43360) INFO 11-19 04:36:45 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.27, Accepted throughput: 41.50 tokens/s, Drafted throughput: 54.90 tokens/s, Accepted: 415 tokens, Drafted: 549 tokens, Per-position acceptance rate: 0.896, 0.749, 0.623, Avg Draft acceptance rate: 75.6%
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
300
  INFO: 127.0.0.1:62756 - "POST /v1/completions HTTP/1.1" 200 OK
301
  (APIServer pid=43360) INFO 11-19 04:36:45 [loggers.py:320] Engine 000: Avg prompt throughput: 8.5 tokens/s, Avg generation throughput: 59.7 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
302
  (APIServer pid=43360) INFO 11-19 04:36:45 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.27, Accepted throughput: 41.50 tokens/s, Drafted throughput: 54.90 tokens/s, Accepted: 415 tokens, Drafted: 549 tokens, Per-position acceptance rate: 0.896, 0.749, 0.623, Avg Draft acceptance rate: 75.6%
303
+ INFO: 127.0.0.1:31980 - "POST /v1/completions HTTP/1.1" 200 OK
304
+ (APIServer pid=43360) INFO 11-19 04:36:55 [loggers.py:320] Engine 000: Avg prompt throughput: 3.4 tokens/s, Avg generation throughput: 50.7 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
305
+ (APIServer pid=43360) INFO 11-19 04:36:55 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.77, Accepted throughput: 32.40 tokens/s, Drafted throughput: 54.89 tokens/s, Accepted: 324 tokens, Drafted: 549 tokens, Per-position acceptance rate: 0.760, 0.585, 0.426, Avg Draft acceptance rate: 59.0%
306
+ INFO: 127.0.0.1:41540 - "POST /v1/completions HTTP/1.1" 200 OK
307
+ INFO: 127.0.0.1:33132 - "POST /v1/completions HTTP/1.1" 200 OK
308
+ INFO: 127.0.0.1:22236 - "POST /v1/completions HTTP/1.1" 200 OK
309
+ (APIServer pid=43360) INFO 11-19 04:37:05 [loggers.py:320] Engine 000: Avg prompt throughput: 48.8 tokens/s, Avg generation throughput: 60.5 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
310
+ (APIServer pid=43360) INFO 11-19 04:37:05 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.39, Accepted throughput: 42.69 tokens/s, Drafted throughput: 53.69 tokens/s, Accepted: 427 tokens, Drafted: 537 tokens, Per-position acceptance rate: 0.899, 0.793, 0.693, Avg Draft acceptance rate: 79.5%
311
+ INFO: 127.0.0.1:20593 - "POST /v1/completions HTTP/1.1" 200 OK
312
+ (APIServer pid=43360) INFO 11-19 04:37:15 [loggers.py:320] Engine 000: Avg prompt throughput: 8.1 tokens/s, Avg generation throughput: 62.6 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
313
+ (APIServer pid=43360) INFO 11-19 04:37:15 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.42, Accepted throughput: 44.30 tokens/s, Drafted throughput: 54.90 tokens/s, Accepted: 443 tokens, Drafted: 549 tokens, Per-position acceptance rate: 0.940, 0.803, 0.678, Avg Draft acceptance rate: 80.7%
314
+ INFO: 127.0.0.1:35913 - "POST /v1/completions HTTP/1.1" 200 OK
315
+ (APIServer pid=43360) INFO 11-19 04:37:25 [loggers.py:320] Engine 000: Avg prompt throughput: 2.4 tokens/s, Avg generation throughput: 60.6 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
316
+ (APIServer pid=43360) INFO 11-19 04:37:25 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.31, Accepted throughput: 42.20 tokens/s, Drafted throughput: 54.90 tokens/s, Accepted: 422 tokens, Drafted: 549 tokens, Per-position acceptance rate: 0.896, 0.776, 0.634, Avg Draft acceptance rate: 76.9%
317
+ INFO: 127.0.0.1:27333 - "POST /v1/completions HTTP/1.1" 200 OK
318
+ (APIServer pid=43360) INFO 11-19 04:37:35 [loggers.py:320] Engine 000: Avg prompt throughput: 3.4 tokens/s, Avg generation throughput: 46.8 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
319
+ (APIServer pid=43360) INFO 11-19 04:37:35 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.53, Accepted throughput: 28.30 tokens/s, Drafted throughput: 55.50 tokens/s, Accepted: 283 tokens, Drafted: 555 tokens, Per-position acceptance rate: 0.730, 0.497, 0.303, Avg Draft acceptance rate: 51.0%
320
+ INFO: 127.0.0.1:41930 - "POST /v1/completions HTTP/1.1" 200 OK
321
+ INFO: 127.0.0.1:54533 - "POST /v1/completions HTTP/1.1" 200 OK
322
+ (APIServer pid=43360) INFO 11-19 04:37:45 [loggers.py:320] Engine 000: Avg prompt throughput: 9.0 tokens/s, Avg generation throughput: 47.4 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
323
+ (APIServer pid=43360) INFO 11-19 04:37:45 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.62, Accepted throughput: 29.30 tokens/s, Drafted throughput: 54.29 tokens/s, Accepted: 293 tokens, Drafted: 543 tokens, Per-position acceptance rate: 0.762, 0.514, 0.343, Avg Draft acceptance rate: 54.0%
324
+ INFO: 127.0.0.1:59323 - "POST /v1/completions HTTP/1.1" 200 OK
325
+ INFO: 127.0.0.1:58551 - "POST /v1/completions HTTP/1.1" 200 OK
326
+ (APIServer pid=43360) INFO 11-19 04:37:55 [loggers.py:320] Engine 000: Avg prompt throughput: 12.1 tokens/s, Avg generation throughput: 62.7 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
327
+ (APIServer pid=43360) INFO 11-19 04:37:55 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.46, Accepted throughput: 44.60 tokens/s, Drafted throughput: 54.30 tokens/s, Accepted: 446 tokens, Drafted: 543 tokens, Per-position acceptance rate: 0.912, 0.845, 0.707, Avg Draft acceptance rate: 82.1%
328
+ INFO: 127.0.0.1:64866 - "POST /v1/completions HTTP/1.1" 200 OK
329
+ INFO: 127.0.0.1:36957 - "GET /metrics HTTP/1.1" 200 OK
330
+ INFO: 127.0.0.1:38598 - "GET /metrics HTTP/1.1" 200 OK
331
+ INFO: 127.0.0.1:39741 - "POST /v1/completions HTTP/1.1" 200 OK
332
+ INFO: 127.0.0.1:58844 - "POST /v1/completions HTTP/1.1" 200 OK
333
+ (APIServer pid=43360) INFO 11-19 04:38:05 [loggers.py:320] Engine 000: Avg prompt throughput: 37.6 tokens/s, Avg generation throughput: 64.5 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
334
+ (APIServer pid=43360) INFO 11-19 04:38:05 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.60, Accepted throughput: 46.59 tokens/s, Drafted throughput: 53.69 tokens/s, Accepted: 466 tokens, Drafted: 537 tokens, Per-position acceptance rate: 0.922, 0.866, 0.816, Avg Draft acceptance rate: 86.8%
335
+ INFO: 127.0.0.1:31380 - "POST /v1/completions HTTP/1.1" 200 OK
336
+ INFO: 127.0.0.1:61154 - "POST /v1/completions HTTP/1.1" 200 OK
337
+ (APIServer pid=43360) INFO 11-19 04:38:15 [loggers.py:320] Engine 000: Avg prompt throughput: 33.2 tokens/s, Avg generation throughput: 68.1 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
338
+ (APIServer pid=43360) INFO 11-19 04:38:15 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.78, Accepted throughput: 49.99 tokens/s, Drafted throughput: 53.99 tokens/s, Accepted: 500 tokens, Drafted: 540 tokens, Per-position acceptance rate: 0.983, 0.922, 0.872, Avg Draft acceptance rate: 92.6%
339
+ INFO: 127.0.0.1:40154 - "POST /v1/completions HTTP/1.1" 200 OK
340
+ INFO: 127.0.0.1:60278 - "POST /v1/completions HTTP/1.1" 200 OK
341
+ (APIServer pid=43360) INFO 11-19 04:38:25 [loggers.py:320] Engine 000: Avg prompt throughput: 50.7 tokens/s, Avg generation throughput: 71.1 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
342
+ (APIServer pid=43360) INFO 11-19 04:38:25 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.97, Accepted throughput: 53.39 tokens/s, Drafted throughput: 53.99 tokens/s, Accepted: 534 tokens, Drafted: 540 tokens, Per-position acceptance rate: 1.000, 0.983, 0.983, Avg Draft acceptance rate: 98.9%
343
+ INFO: 127.0.0.1:26809 - "POST /v1/completions HTTP/1.1" 200 OK
344
+ INFO: 127.0.0.1:64009 - "POST /v1/completions HTTP/1.1" 200 OK
345
+ (APIServer pid=43360) INFO 11-19 04:38:35 [loggers.py:320] Engine 000: Avg prompt throughput: 35.0 tokens/s, Avg generation throughput: 68.2 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
346
+ (APIServer pid=43360) INFO 11-19 04:38:35 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.79, Accepted throughput: 50.50 tokens/s, Drafted throughput: 54.30 tokens/s, Accepted: 505 tokens, Drafted: 543 tokens, Per-position acceptance rate: 0.967, 0.928, 0.895, Avg Draft acceptance rate: 93.0%
347
+ INFO: 127.0.0.1:65045 - "POST /v1/completions HTTP/1.1" 200 OK
348
+ INFO: 127.0.0.1:53505 - "POST /v1/completions HTTP/1.1" 200 OK
349
+ INFO: 127.0.0.1:35796 - "POST /v1/completions HTTP/1.1" 200 OK
350
+ INFO: 127.0.0.1:20738 - "POST /v1/completions HTTP/1.1" 200 OK
351
+ (APIServer pid=43360) INFO 11-19 04:38:45 [loggers.py:320] Engine 000: Avg prompt throughput: 43.9 tokens/s, Avg generation throughput: 65.8 tokens/s, Running: 0 reqs, Waiting: 0 reqs, GPU KV cache usage: 0.0%
352
+ (APIServer pid=43360) INFO 11-19 04:38:45 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.72, Accepted throughput: 48.20 tokens/s, Drafted throughput: 53.09 tokens/s, Accepted: 482 tokens, Drafted: 531 tokens, Per-position acceptance rate: 0.949, 0.910, 0.864, Avg Draft acceptance rate: 90.8%
353
+ INFO: 127.0.0.1:24832 - "POST /v1/completions HTTP/1.1" 200 OK
354
+ (APIServer pid=43360) INFO 11-19 04:38:55 [loggers.py:320] Engine 000: Avg prompt throughput: 55.2 tokens/s, Avg generation throughput: 71.0 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
355
+ (APIServer pid=43360) INFO 11-19 04:38:55 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.91, Accepted throughput: 52.70 tokens/s, Drafted throughput: 54.30 tokens/s, Accepted: 527 tokens, Drafted: 543 tokens, Per-position acceptance rate: 0.989, 0.972, 0.950, Avg Draft acceptance rate: 97.1%
356
+ INFO: 127.0.0.1:49785 - "POST /v1/completions HTTP/1.1" 200 OK
357
+ INFO: 127.0.0.1:57626 - "POST /v1/completions HTTP/1.1" 200 OK
358
+ INFO: 127.0.0.1:32001 - "POST /v1/completions HTTP/1.1" 200 OK
359
+ (APIServer pid=43360) INFO 11-19 04:39:05 [loggers.py:320] Engine 000: Avg prompt throughput: 71.0 tokens/s, Avg generation throughput: 65.3 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
360
+ (APIServer pid=43360) INFO 11-19 04:39:05 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.67, Accepted throughput: 47.49 tokens/s, Drafted throughput: 53.39 tokens/s, Accepted: 475 tokens, Drafted: 534 tokens, Per-position acceptance rate: 0.949, 0.899, 0.820, Avg Draft acceptance rate: 89.0%
361
+ INFO: 127.0.0.1:21162 - "POST /v1/completions HTTP/1.1" 200 OK
362
+ (APIServer pid=43360) INFO 11-19 04:39:15 [loggers.py:320] Engine 000: Avg prompt throughput: 17.5 tokens/s, Avg generation throughput: 69.4 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
363
+ (APIServer pid=43360) INFO 11-19 04:39:15 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.80, Accepted throughput: 51.29 tokens/s, Drafted throughput: 54.89 tokens/s, Accepted: 513 tokens, Drafted: 549 tokens, Per-position acceptance rate: 0.962, 0.951, 0.891, Avg Draft acceptance rate: 93.4%
364
+ INFO: 127.0.0.1:45689 - "POST /v1/completions HTTP/1.1" 200 OK
365
+ INFO: 127.0.0.1:43576 - "POST /v1/completions HTTP/1.1" 200 OK
366
+ INFO: 127.0.0.1:32990 - "POST /v1/completions HTTP/1.1" 200 OK
367
+ (APIServer pid=43360) INFO 11-19 04:39:25 [loggers.py:320] Engine 000: Avg prompt throughput: 47.8 tokens/s, Avg generation throughput: 65.5 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
368
+ (APIServer pid=43360) INFO 11-19 04:39:25 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.69, Accepted throughput: 47.79 tokens/s, Drafted throughput: 53.39 tokens/s, Accepted: 478 tokens, Drafted: 534 tokens, Per-position acceptance rate: 0.966, 0.893, 0.826, Avg Draft acceptance rate: 89.5%
369
+ INFO: 127.0.0.1:55943 - "POST /v1/completions HTTP/1.1" 200 OK
370
+ INFO: 127.0.0.1:21227 - "POST /v1/completions HTTP/1.1" 200 OK
371
+ INFO: 127.0.0.1:29243 - "POST /v1/completions HTTP/1.1" 200 OK
372
+ (APIServer pid=43360) INFO 11-19 04:39:35 [loggers.py:320] Engine 000: Avg prompt throughput: 53.8 tokens/s, Avg generation throughput: 70.4 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
373
+ (APIServer pid=43360) INFO 11-19 04:39:35 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.95, Accepted throughput: 52.79 tokens/s, Drafted throughput: 53.69 tokens/s, Accepted: 528 tokens, Drafted: 537 tokens, Per-position acceptance rate: 0.994, 0.983, 0.972, Avg Draft acceptance rate: 98.3%
374
+ INFO: 127.0.0.1:33014 - "POST /v1/completions HTTP/1.1" 200 OK
375
+ (APIServer pid=43360) INFO 11-19 04:39:45 [loggers.py:320] Engine 000: Avg prompt throughput: 16.7 tokens/s, Avg generation throughput: 59.4 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
376
+ (APIServer pid=43360) INFO 11-19 04:39:45 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.27, Accepted throughput: 41.39 tokens/s, Drafted throughput: 54.59 tokens/s, Accepted: 414 tokens, Drafted: 546 tokens, Per-position acceptance rate: 0.885, 0.758, 0.632, Avg Draft acceptance rate: 75.8%
377
+ INFO: 127.0.0.1:51985 - "POST /v1/completions HTTP/1.1" 200 OK
378
+ INFO: 127.0.0.1:62023 - "POST /v1/completions HTTP/1.1" 200 OK
379
+ INFO: 127.0.0.1:58405 - "POST /v1/completions HTTP/1.1" 200 OK
380
+ (APIServer pid=43360) INFO 11-19 04:39:55 [loggers.py:320] Engine 000: Avg prompt throughput: 56.0 tokens/s, Avg generation throughput: 66.5 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
381
+ (APIServer pid=43360) INFO 11-19 04:39:55 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.72, Accepted throughput: 48.69 tokens/s, Drafted throughput: 53.69 tokens/s, Accepted: 487 tokens, Drafted: 537 tokens, Per-position acceptance rate: 0.955, 0.888, 0.877, Avg Draft acceptance rate: 90.7%
382
+ INFO: 127.0.0.1:62117 - "POST /v1/completions HTTP/1.1" 200 OK
383
+ INFO: 127.0.0.1:54736 - "POST /v1/completions HTTP/1.1" 200 OK
384
+ INFO: 127.0.0.1:19213 - "POST /v1/completions HTTP/1.1" 200 OK
385
+ INFO: 127.0.0.1:30545 - "POST /v1/completions HTTP/1.1" 200 OK
386
+ (APIServer pid=43360) INFO 11-19 04:40:05 [loggers.py:320] Engine 000: Avg prompt throughput: 79.2 tokens/s, Avg generation throughput: 68.3 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
387
+ (APIServer pid=43360) INFO 11-19 04:40:05 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.90, Accepted throughput: 50.99 tokens/s, Drafted throughput: 52.79 tokens/s, Accepted: 510 tokens, Drafted: 528 tokens, Per-position acceptance rate: 0.989, 0.972, 0.938, Avg Draft acceptance rate: 96.6%
388
+ INFO: 127.0.0.1:60183 - "POST /v1/completions HTTP/1.1" 200 OK
389
+ INFO: 127.0.0.1:20441 - "POST /v1/completions HTTP/1.1" 200 OK
390
+ (APIServer pid=43360) INFO 11-19 04:40:15 [loggers.py:320] Engine 000: Avg prompt throughput: 26.6 tokens/s, Avg generation throughput: 66.5 tokens/s, Running: 0 reqs, Waiting: 0 reqs, GPU KV cache usage: 0.0%
391
+ (APIServer pid=43360) INFO 11-19 04:40:15 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.68, Accepted throughput: 48.80 tokens/s, Drafted throughput: 54.60 tokens/s, Accepted: 488 tokens, Drafted: 546 tokens, Per-position acceptance rate: 0.934, 0.896, 0.852, Avg Draft acceptance rate: 89.4%
392
+ INFO: 127.0.0.1:19913 - "POST /v1/completions HTTP/1.1" 200 OK
393
+ INFO: 127.0.0.1:48460 - "POST /v1/completions HTTP/1.1" 200 OK
394
+ INFO: 127.0.0.1:58257 - "POST /v1/completions HTTP/1.1" 200 OK
395
+ (APIServer pid=43360) INFO 11-19 04:40:25 [loggers.py:320] Engine 000: Avg prompt throughput: 71.7 tokens/s, Avg generation throughput: 69.3 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
396
+ (APIServer pid=43360) INFO 11-19 04:40:25 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.94, Accepted throughput: 52.00 tokens/s, Drafted throughput: 53.10 tokens/s, Accepted: 520 tokens, Drafted: 531 tokens, Per-position acceptance rate: 1.000, 0.989, 0.949, Avg Draft acceptance rate: 97.9%
397
+ INFO: 127.0.0.1:64507 - "POST /v1/completions HTTP/1.1" 200 OK
398
+ (APIServer pid=43360) INFO 11-19 04:40:35 [loggers.py:320] Engine 000: Avg prompt throughput: 20.8 tokens/s, Avg generation throughput: 65.7 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
399
+ (APIServer pid=43360) INFO 11-19 04:40:35 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.61, Accepted throughput: 47.50 tokens/s, Drafted throughput: 54.60 tokens/s, Accepted: 475 tokens, Drafted: 546 tokens, Per-position acceptance rate: 0.940, 0.868, 0.802, Avg Draft acceptance rate: 87.0%
400
+ INFO: 127.0.0.1:34883 - "POST /v1/completions HTTP/1.1" 200 OK
401
+ INFO: 127.0.0.1:19156 - "POST /v1/completions HTTP/1.1" 200 OK
402
+ INFO: 127.0.0.1:26441 - "POST /v1/completions HTTP/1.1" 200 OK
403
+ INFO: 127.0.0.1:45621 - "GET /metrics HTTP/1.1" 200 OK
404
+ INFO: 127.0.0.1:33052 - "GET /metrics HTTP/1.1" 200 OK
405
+ INFO: 127.0.0.1:18501 - "POST /v1/completions HTTP/1.1" 200 OK
406
+ (APIServer pid=43360) INFO 11-19 04:40:45 [loggers.py:320] Engine 000: Avg prompt throughput: 52.1 tokens/s, Avg generation throughput: 63.3 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
407
+ (APIServer pid=43360) INFO 11-19 04:40:45 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.56, Accepted throughput: 45.40 tokens/s, Drafted throughput: 53.10 tokens/s, Accepted: 454 tokens, Drafted: 531 tokens, Per-position acceptance rate: 0.921, 0.842, 0.802, Avg Draft acceptance rate: 85.5%
408
+ INFO: 127.0.0.1:54679 - "POST /v1/completions HTTP/1.1" 200 OK
409
+ (APIServer pid=43360) INFO 11-19 04:40:55 [loggers.py:320] Engine 000: Avg prompt throughput: 12.1 tokens/s, Avg generation throughput: 69.6 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
410
+ (APIServer pid=43360) INFO 11-19 04:40:55 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.79, Accepted throughput: 51.29 tokens/s, Drafted throughput: 55.19 tokens/s, Accepted: 513 tokens, Drafted: 552 tokens, Per-position acceptance rate: 0.989, 0.940, 0.859, Avg Draft acceptance rate: 92.9%
411
+ INFO: 127.0.0.1:28450 - "POST /v1/completions HTTP/1.1" 200 OK
412
+ INFO: 127.0.0.1:42844 - "POST /v1/completions HTTP/1.1" 200 OK
413
+ INFO: 127.0.0.1:50578 - "POST /v1/completions HTTP/1.1" 200 OK
414
+ (APIServer pid=43360) INFO 11-19 04:41:05 [loggers.py:320] Engine 000: Avg prompt throughput: 19.1 tokens/s, Avg generation throughput: 64.2 tokens/s, Running: 0 reqs, Waiting: 0 reqs, GPU KV cache usage: 0.0%
415
+ (APIServer pid=43360) INFO 11-19 04:41:05 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.59, Accepted throughput: 46.59 tokens/s, Drafted throughput: 53.99 tokens/s, Accepted: 466 tokens, Drafted: 540 tokens, Per-position acceptance rate: 0.967, 0.850, 0.772, Avg Draft acceptance rate: 86.3%
416
+ INFO: 127.0.0.1:17385 - "POST /v1/completions HTTP/1.1" 200 OK
417
+ INFO: 127.0.0.1:35359 - "POST /v1/completions HTTP/1.1" 200 OK
418
+ (APIServer pid=43360) INFO 11-19 04:41:15 [loggers.py:320] Engine 000: Avg prompt throughput: 31.4 tokens/s, Avg generation throughput: 63.9 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
419
+ (APIServer pid=43360) INFO 11-19 04:41:15 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.55, Accepted throughput: 45.89 tokens/s, Drafted throughput: 53.99 tokens/s, Accepted: 459 tokens, Drafted: 540 tokens, Per-position acceptance rate: 0.933, 0.850, 0.767, Avg Draft acceptance rate: 85.0%
420
+ INFO: 127.0.0.1:41955 - "POST /v1/completions HTTP/1.1" 200 OK
421
+ INFO: 127.0.0.1:40825 - "POST /v1/completions HTTP/1.1" 200 OK
422
+ (APIServer pid=43360) INFO 11-19 04:41:25 [loggers.py:320] Engine 000: Avg prompt throughput: 23.3 tokens/s, Avg generation throughput: 68.0 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
423
+ (APIServer pid=43360) INFO 11-19 04:41:25 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.75, Accepted throughput: 49.80 tokens/s, Drafted throughput: 54.30 tokens/s, Accepted: 498 tokens, Drafted: 543 tokens, Per-position acceptance rate: 0.967, 0.928, 0.856, Avg Draft acceptance rate: 91.7%
424
+ INFO: 127.0.0.1:21633 - "POST /v1/completions HTTP/1.1" 200 OK
425
+ INFO: 127.0.0.1:54753 - "POST /v1/completions HTTP/1.1" 200 OK
426
+ (APIServer pid=43360) INFO 11-19 04:41:35 [loggers.py:320] Engine 000: Avg prompt throughput: 22.0 tokens/s, Avg generation throughput: 67.2 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
427
+ (APIServer pid=43360) INFO 11-19 04:41:35 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.71, Accepted throughput: 49.29 tokens/s, Drafted throughput: 54.59 tokens/s, Accepted: 493 tokens, Drafted: 546 tokens, Per-position acceptance rate: 0.978, 0.907, 0.824, Avg Draft acceptance rate: 90.3%
428
+ INFO: 127.0.0.1:39533 - "POST /v1/completions HTTP/1.1" 200 OK
429
+ INFO: 127.0.0.1:64541 - "POST /v1/completions HTTP/1.1" 200 OK
430
+ (APIServer pid=43360) INFO 11-19 04:41:45 [loggers.py:320] Engine 000: Avg prompt throughput: 19.7 tokens/s, Avg generation throughput: 64.8 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
431
+ (APIServer pid=43360) INFO 11-19 04:41:45 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.59, Accepted throughput: 46.89 tokens/s, Drafted throughput: 54.29 tokens/s, Accepted: 469 tokens, Drafted: 543 tokens, Per-position acceptance rate: 0.945, 0.867, 0.779, Avg Draft acceptance rate: 86.4%
432
+ INFO: 127.0.0.1:37297 - "POST /v1/completions HTTP/1.1" 200 OK
433
+ INFO: 127.0.0.1:40472 - "POST /v1/completions HTTP/1.1" 200 OK
434
+ (APIServer pid=43360) INFO 11-19 04:41:55 [loggers.py:320] Engine 000: Avg prompt throughput: 24.0 tokens/s, Avg generation throughput: 67.6 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
435
+ (APIServer pid=43360) INFO 11-19 04:41:55 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.74, Accepted throughput: 49.59 tokens/s, Drafted throughput: 54.29 tokens/s, Accepted: 496 tokens, Drafted: 543 tokens, Per-position acceptance rate: 0.972, 0.923, 0.845, Avg Draft acceptance rate: 91.3%
436
+ INFO: 127.0.0.1:61726 - "POST /v1/completions HTTP/1.1" 200 OK
437
+ INFO: 127.0.0.1:31130 - "POST /v1/completions HTTP/1.1" 200 OK
438
+ (APIServer pid=43360) INFO 11-19 04:42:05 [loggers.py:320] Engine 000: Avg prompt throughput: 23.6 tokens/s, Avg generation throughput: 66.7 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
439
+ (APIServer pid=43360) INFO 11-19 04:42:05 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.69, Accepted throughput: 48.69 tokens/s, Drafted throughput: 54.29 tokens/s, Accepted: 487 tokens, Drafted: 543 tokens, Per-position acceptance rate: 0.956, 0.906, 0.829, Avg Draft acceptance rate: 89.7%
440
+ INFO: 127.0.0.1:38154 - "POST /v1/completions HTTP/1.1" 200 OK
441
+ INFO: 127.0.0.1:63291 - "POST /v1/completions HTTP/1.1" 200 OK
442
+ (APIServer pid=43360) INFO 11-19 04:42:15 [loggers.py:320] Engine 000: Avg prompt throughput: 21.1 tokens/s, Avg generation throughput: 68.1 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
443
+ (APIServer pid=43360) INFO 11-19 04:42:15 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.77, Accepted throughput: 50.09 tokens/s, Drafted throughput: 54.29 tokens/s, Accepted: 501 tokens, Drafted: 543 tokens, Per-position acceptance rate: 0.972, 0.917, 0.878, Avg Draft acceptance rate: 92.3%
444
+ INFO: 127.0.0.1:47205 - "POST /v1/completions HTTP/1.1" 200 OK
445
+ INFO: 127.0.0.1:37525 - "POST /v1/completions HTTP/1.1" 200 OK
446
+ INFO: 127.0.0.1:47535 - "POST /v1/completions HTTP/1.1" 200 OK
447
+ (APIServer pid=43360) INFO 11-19 04:42:25 [loggers.py:320] Engine 000: Avg prompt throughput: 26.7 tokens/s, Avg generation throughput: 63.9 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
448
+ (APIServer pid=43360) INFO 11-19 04:42:25 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.60, Accepted throughput: 46.30 tokens/s, Drafted throughput: 53.40 tokens/s, Accepted: 463 tokens, Drafted: 534 tokens, Per-position acceptance rate: 0.938, 0.871, 0.792, Avg Draft acceptance rate: 86.7%
449
+ INFO: 127.0.0.1:29365 - "POST /v1/completions HTTP/1.1" 200 OK
450
+ (APIServer pid=43360) INFO 11-19 04:42:35 [loggers.py:320] Engine 000: Avg prompt throughput: 9.8 tokens/s, Avg generation throughput: 69.8 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
451
+ (APIServer pid=43360) INFO 11-19 04:42:35 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.80, Accepted throughput: 51.50 tokens/s, Drafted throughput: 55.20 tokens/s, Accepted: 515 tokens, Drafted: 552 tokens, Per-position acceptance rate: 0.989, 0.924, 0.886, Avg Draft acceptance rate: 93.3%
452
+ INFO: 127.0.0.1:53224 - "POST /v1/completions HTTP/1.1" 200 OK
453
+ INFO: 127.0.0.1:60452 - "POST /v1/completions HTTP/1.1" 200 OK
454
+ (APIServer pid=43360) INFO 11-19 04:42:45 [loggers.py:320] Engine 000: Avg prompt throughput: 17.8 tokens/s, Avg generation throughput: 66.1 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
455
+ (APIServer pid=43360) INFO 11-19 04:42:45 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.67, Accepted throughput: 48.00 tokens/s, Drafted throughput: 54.00 tokens/s, Accepted: 480 tokens, Drafted: 540 tokens, Per-position acceptance rate: 0.956, 0.906, 0.806, Avg Draft acceptance rate: 88.9%
456
+ INFO: 127.0.0.1:53722 - "POST /v1/completions HTTP/1.1" 200 OK
457
+ INFO: 127.0.0.1:22977 - "POST /v1/completions HTTP/1.1" 200 OK
458
+ (APIServer pid=43360) INFO 11-19 04:42:55 [loggers.py:320] Engine 000: Avg prompt throughput: 20.3 tokens/s, Avg generation throughput: 65.4 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
459
+ (APIServer pid=43360) INFO 11-19 04:42:55 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.62, Accepted throughput: 47.60 tokens/s, Drafted throughput: 54.60 tokens/s, Accepted: 476 tokens, Drafted: 546 tokens, Per-position acceptance rate: 0.951, 0.868, 0.797, Avg Draft acceptance rate: 87.2%
460
+ INFO: 127.0.0.1:51080 - "POST /v1/completions HTTP/1.1" 200 OK
461
+ INFO: 127.0.0.1:21374 - "POST /v1/completions HTTP/1.1" 200 OK
462
+ (APIServer pid=43360) INFO 11-19 04:43:05 [loggers.py:320] Engine 000: Avg prompt throughput: 14.6 tokens/s, Avg generation throughput: 68.5 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
463
+ (APIServer pid=43360) INFO 11-19 04:43:05 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.80, Accepted throughput: 50.69 tokens/s, Drafted throughput: 54.29 tokens/s, Accepted: 507 tokens, Drafted: 543 tokens, Per-position acceptance rate: 0.972, 0.928, 0.901, Avg Draft acceptance rate: 93.4%
464
+ INFO: 127.0.0.1:49837 - "POST /v1/completions HTTP/1.1" 200 OK
465
+ INFO: 127.0.0.1:22927 - "POST /v1/completions HTTP/1.1" 200 OK
466
+ INFO: 127.0.0.1:24599 - "POST /v1/completions HTTP/1.1" 200 OK
467
+ (APIServer pid=43360) INFO 11-19 04:43:15 [loggers.py:320] Engine 000: Avg prompt throughput: 17.0 tokens/s, Avg generation throughput: 64.9 tokens/s, Running: 0 reqs, Waiting: 0 reqs, GPU KV cache usage: 0.0%
468
+ (APIServer pid=43360) INFO 11-19 04:43:15 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.59, Accepted throughput: 46.89 tokens/s, Drafted throughput: 54.29 tokens/s, Accepted: 469 tokens, Drafted: 543 tokens, Per-position acceptance rate: 0.950, 0.862, 0.779, Avg Draft acceptance rate: 86.4%
469
+ INFO: 127.0.0.1:57270 - "POST /v1/completions HTTP/1.1" 200 OK
470
+ INFO: 127.0.0.1:32418 - "POST /v1/completions HTTP/1.1" 200 OK
471
+ INFO: 127.0.0.1:34390 - "POST /v1/completions HTTP/1.1" 200 OK
472
+ (APIServer pid=43360) INFO 11-19 04:43:25 [loggers.py:320] Engine 000: Avg prompt throughput: 30.2 tokens/s, Avg generation throughput: 68.0 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
473
+ (APIServer pid=43360) INFO 11-19 04:43:25 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.84, Accepted throughput: 50.20 tokens/s, Drafted throughput: 53.10 tokens/s, Accepted: 502 tokens, Drafted: 531 tokens, Per-position acceptance rate: 0.989, 0.944, 0.904, Avg Draft acceptance rate: 94.5%
474
+ INFO: 127.0.0.1:26840 - "POST /v1/completions HTTP/1.1" 200 OK
475
+ INFO: 127.0.0.1:17413 - "POST /v1/completions HTTP/1.1" 200 OK
476
+ (APIServer pid=43360) INFO 11-19 04:43:35 [loggers.py:320] Engine 000: Avg prompt throughput: 19.7 tokens/s, Avg generation throughput: 69.6 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
477
+ (APIServer pid=43360) INFO 11-19 04:43:35 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.83, Accepted throughput: 51.29 tokens/s, Drafted throughput: 54.29 tokens/s, Accepted: 513 tokens, Drafted: 543 tokens, Per-position acceptance rate: 0.978, 0.967, 0.890, Avg Draft acceptance rate: 94.5%
478
+ INFO: 127.0.0.1:36820 - "POST /v1/completions HTTP/1.1" 200 OK
479
+ INFO: 127.0.0.1:46315 - "POST /v1/completions HTTP/1.1" 200 OK
480
+ INFO: 127.0.0.1:25401 - "POST /v1/completions HTTP/1.1" 200 OK
481
+ (APIServer pid=43360) INFO 11-19 04:43:45 [loggers.py:320] Engine 000: Avg prompt throughput: 25.2 tokens/s, Avg generation throughput: 64.8 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
482
+ (APIServer pid=43360) INFO 11-19 04:43:45 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.62, Accepted throughput: 46.89 tokens/s, Drafted throughput: 53.69 tokens/s, Accepted: 469 tokens, Drafted: 537 tokens, Per-position acceptance rate: 0.955, 0.899, 0.765, Avg Draft acceptance rate: 87.3%
483
+ INFO: 127.0.0.1:41672 - "GET /metrics HTTP/1.1" 200 OK
484
+ INFO: 127.0.0.1:64769 - "GET /metrics HTTP/1.1" 200 OK
485
+ INFO: 127.0.0.1:51498 - "POST /v1/completions HTTP/1.1" 200 OK
486
+ (APIServer pid=43360) INFO 11-19 04:43:55 [loggers.py:320] Engine 000: Avg prompt throughput: 4.2 tokens/s, Avg generation throughput: 54.2 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
487
+ (APIServer pid=43360) INFO 11-19 04:43:55 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.95, Accepted throughput: 35.89 tokens/s, Drafted throughput: 55.19 tokens/s, Accepted: 359 tokens, Drafted: 552 tokens, Per-position acceptance rate: 0.788, 0.641, 0.522, Avg Draft acceptance rate: 65.0%
488
+ (APIServer pid=43360) INFO 11-19 04:44:05 [loggers.py:320] Engine 000: Avg prompt throughput: 0.0 tokens/s, Avg generation throughput: 51.8 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.4%
489
+ (APIServer pid=43360) INFO 11-19 04:44:05 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.82, Accepted throughput: 33.40 tokens/s, Drafted throughput: 55.20 tokens/s, Accepted: 334 tokens, Drafted: 552 tokens, Per-position acceptance rate: 0.837, 0.592, 0.386, Avg Draft acceptance rate: 60.5%
490
+ INFO: 127.0.0.1:55381 - "POST /v1/completions HTTP/1.1" 200 OK
491
+ (APIServer pid=43360) INFO 11-19 04:44:15 [loggers.py:320] Engine 000: Avg prompt throughput: 3.8 tokens/s, Avg generation throughput: 46.3 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
492
+ (APIServer pid=43360) INFO 11-19 04:44:15 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.54, Accepted throughput: 28.10 tokens/s, Drafted throughput: 54.90 tokens/s, Accepted: 281 tokens, Drafted: 549 tokens, Per-position acceptance rate: 0.743, 0.497, 0.295, Avg Draft acceptance rate: 51.2%
493
+ (APIServer pid=43360) INFO 11-19 04:44:25 [loggers.py:320] Engine 000: Avg prompt throughput: 0.0 tokens/s, Avg generation throughput: 48.3 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.4%
494
+ (APIServer pid=43360) INFO 11-19 04:44:25 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.62, Accepted throughput: 29.90 tokens/s, Drafted throughput: 55.20 tokens/s, Accepted: 299 tokens, Drafted: 552 tokens, Per-position acceptance rate: 0.772, 0.516, 0.337, Avg Draft acceptance rate: 54.2%
495
+ INFO: 127.0.0.1:50261 - "POST /v1/completions HTTP/1.1" 200 OK
496
+ (APIServer pid=43360) INFO 11-19 04:44:35 [loggers.py:320] Engine 000: Avg prompt throughput: 4.2 tokens/s, Avg generation throughput: 43.8 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
497
+ (APIServer pid=43360) INFO 11-19 04:44:35 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.38, Accepted throughput: 25.30 tokens/s, Drafted throughput: 55.19 tokens/s, Accepted: 253 tokens, Drafted: 552 tokens, Per-position acceptance rate: 0.712, 0.413, 0.250, Avg Draft acceptance rate: 45.8%
498
+ (APIServer pid=43360) INFO 11-19 04:44:45 [loggers.py:320] Engine 000: Avg prompt throughput: 0.0 tokens/s, Avg generation throughput: 46.7 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.4%
499
+ (APIServer pid=43360) INFO 11-19 04:44:45 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.54, Accepted throughput: 28.30 tokens/s, Drafted throughput: 55.19 tokens/s, Accepted: 283 tokens, Drafted: 552 tokens, Per-position acceptance rate: 0.728, 0.505, 0.304, Avg Draft acceptance rate: 51.3%
500
+ INFO: 127.0.0.1:29748 - "POST /v1/completions HTTP/1.1" 200 OK
501
+ (APIServer pid=43360) INFO 11-19 04:44:55 [loggers.py:320] Engine 000: Avg prompt throughput: 7.0 tokens/s, Avg generation throughput: 48.6 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
502
+ (APIServer pid=43360) INFO 11-19 04:44:55 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.65, Accepted throughput: 30.20 tokens/s, Drafted throughput: 54.90 tokens/s, Accepted: 302 tokens, Drafted: 549 tokens, Per-position acceptance rate: 0.787, 0.525, 0.339, Avg Draft acceptance rate: 55.0%
503
+ INFO: 127.0.0.1:39459 - "POST /v1/completions HTTP/1.1" 200 OK
504
+ (APIServer pid=43360) INFO 11-19 04:45:05 [loggers.py:320] Engine 000: Avg prompt throughput: 6.7 tokens/s, Avg generation throughput: 48.6 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
505
+ (APIServer pid=43360) INFO 11-19 04:45:05 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.68, Accepted throughput: 30.60 tokens/s, Drafted throughput: 54.60 tokens/s, Accepted: 306 tokens, Drafted: 546 tokens, Per-position acceptance rate: 0.786, 0.555, 0.341, Avg Draft acceptance rate: 56.0%
506
+ (APIServer pid=43360) INFO 11-19 04:45:15 [loggers.py:320] Engine 000: Avg prompt throughput: 0.0 tokens/s, Avg generation throughput: 44.3 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
507
+ (APIServer pid=43360) INFO 11-19 04:45:15 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.39, Accepted throughput: 25.79 tokens/s, Drafted throughput: 55.49 tokens/s, Accepted: 258 tokens, Drafted: 555 tokens, Per-position acceptance rate: 0.730, 0.449, 0.216, Avg Draft acceptance rate: 46.5%
508
+ INFO: 127.0.0.1:16244 - "POST /v1/completions HTTP/1.1" 200 OK
509
+ (APIServer pid=43360) INFO 11-19 04:45:25 [loggers.py:320] Engine 000: Avg prompt throughput: 4.6 tokens/s, Avg generation throughput: 45.9 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
510
+ (APIServer pid=43360) INFO 11-19 04:45:25 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.50, Accepted throughput: 27.50 tokens/s, Drafted throughput: 54.89 tokens/s, Accepted: 275 tokens, Drafted: 549 tokens, Per-position acceptance rate: 0.743, 0.503, 0.257, Avg Draft acceptance rate: 50.1%
511
+ (APIServer pid=43360) INFO 11-19 04:45:35 [loggers.py:320] Engine 000: Avg prompt throughput: 0.0 tokens/s, Avg generation throughput: 46.1 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
512
+ (APIServer pid=43360) INFO 11-19 04:45:35 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.49, Accepted throughput: 27.60 tokens/s, Drafted throughput: 55.49 tokens/s, Accepted: 276 tokens, Drafted: 555 tokens, Per-position acceptance rate: 0.762, 0.465, 0.265, Avg Draft acceptance rate: 49.7%
513
+ INFO: 127.0.0.1:46544 - "POST /v1/completions HTTP/1.1" 200 OK
514
+ (APIServer pid=43360) INFO 11-19 04:45:45 [loggers.py:320] Engine 000: Avg prompt throughput: 4.3 tokens/s, Avg generation throughput: 46.8 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
515
+ (APIServer pid=43360) INFO 11-19 04:45:45 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.58, Accepted throughput: 28.69 tokens/s, Drafted throughput: 54.59 tokens/s, Accepted: 287 tokens, Drafted: 546 tokens, Per-position acceptance rate: 0.753, 0.478, 0.346, Avg Draft acceptance rate: 52.6%
516
+ (APIServer pid=43360) INFO 11-19 04:45:55 [loggers.py:320] Engine 000: Avg prompt throughput: 0.0 tokens/s, Avg generation throughput: 41.7 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
517
+ (APIServer pid=43360) INFO 11-19 04:45:55 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.25, Accepted throughput: 23.20 tokens/s, Drafted throughput: 55.49 tokens/s, Accepted: 232 tokens, Drafted: 555 tokens, Per-position acceptance rate: 0.670, 0.362, 0.222, Avg Draft acceptance rate: 41.8%
518
+ INFO: 127.0.0.1:53018 - "POST /v1/completions HTTP/1.1" 200 OK
519
+ (APIServer pid=43360) INFO 11-19 04:46:05 [loggers.py:320] Engine 000: Avg prompt throughput: 7.0 tokens/s, Avg generation throughput: 46.7 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
520
+ (APIServer pid=43360) INFO 11-19 04:46:05 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.58, Accepted throughput: 28.70 tokens/s, Drafted throughput: 54.59 tokens/s, Accepted: 287 tokens, Drafted: 546 tokens, Per-position acceptance rate: 0.736, 0.500, 0.341, Avg Draft acceptance rate: 52.6%
521
+ (APIServer pid=43360) INFO 11-19 04:46:15 [loggers.py:320] Engine 000: Avg prompt throughput: 0.0 tokens/s, Avg generation throughput: 42.5 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
522
+ (APIServer pid=43360) INFO 11-19 04:46:15 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.28, Accepted throughput: 23.90 tokens/s, Drafted throughput: 55.79 tokens/s, Accepted: 239 tokens, Drafted: 558 tokens, Per-position acceptance rate: 0.710, 0.371, 0.204, Avg Draft acceptance rate: 42.8%
523
+ INFO: 127.0.0.1:30654 - "POST /v1/completions HTTP/1.1" 200 OK
524
+ (APIServer pid=43360) INFO 11-19 04:46:25 [loggers.py:320] Engine 000: Avg prompt throughput: 3.7 tokens/s, Avg generation throughput: 41.3 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
525
+ (APIServer pid=43360) INFO 11-19 04:46:25 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.27, Accepted throughput: 23.20 tokens/s, Drafted throughput: 54.60 tokens/s, Accepted: 232 tokens, Drafted: 546 tokens, Per-position acceptance rate: 0.698, 0.385, 0.192, Avg Draft acceptance rate: 42.5%
526
+ (APIServer pid=43360) INFO 11-19 04:46:35 [loggers.py:320] Engine 000: Avg prompt throughput: 0.0 tokens/s, Avg generation throughput: 45.0 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
527
+ (APIServer pid=43360) INFO 11-19 04:46:35 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.43, Accepted throughput: 26.50 tokens/s, Drafted throughput: 55.50 tokens/s, Accepted: 265 tokens, Drafted: 555 tokens, Per-position acceptance rate: 0.735, 0.481, 0.216, Avg Draft acceptance rate: 47.7%
528
+ INFO: 127.0.0.1:24854 - "POST /v1/completions HTTP/1.1" 200 OK
529
+ (APIServer pid=43360) INFO 11-19 04:46:45 [loggers.py:320] Engine 000: Avg prompt throughput: 7.2 tokens/s, Avg generation throughput: 46.4 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
530
+ (APIServer pid=43360) INFO 11-19 04:46:45 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.54, Accepted throughput: 28.10 tokens/s, Drafted throughput: 54.59 tokens/s, Accepted: 281 tokens, Drafted: 546 tokens, Per-position acceptance rate: 0.813, 0.456, 0.275, Avg Draft acceptance rate: 51.5%
runs/r3/logs/eval_prev.log CHANGED
@@ -9,3 +9,6 @@ PASS
9
  [04:30:28] repeat 1/1: fresh vLLM process
10
  [04:30:28] starting vLLM (log: /workspace/emc/work/logs/20261119-043028-eval-r1.log)
11
  [04:32:26] vLLM ready after 118s
 
 
 
 
9
  [04:30:28] repeat 1/1: fresh vLLM process
10
  [04:30:28] starting vLLM (log: /workspace/emc/work/logs/20261119-043028-eval-r1.log)
11
  [04:32:26] vLLM ready after 118s
12
+ [04:38:03] chat: 40 prompts done
13
+ [04:40:45] code: 40 prompts done
14
+ [04:43:49] math: 40 prompts done
runs/r3/logs/stats.log CHANGED
@@ -58,3 +58,4 @@
58
  [2026-11-19 04:30:13] ram 0.2/172 GB used (cache 0.0, shmem 0.0) | cgroup 0.2/172 GB | disk 143 GB used | gpu 0.0/95.6 GiB, 0%, 62 W | load 0.0 | largest python 0.1 GB, python 0.1 GB, python 0.0 GB
59
  [2026-11-19 04:30:14] ram 0.2/172 GB used (cache 0.0, shmem 0.0) | cgroup 0.2/172 GB | disk 143 GB used | gpu 0.0/95.6 GiB, 0%, 62 W | load 0.0 | largest python 0.1 GB, python 0.1 GB, python 0.0 GB
60
  [2026-11-19 04:36:46] ram 4.6/172 GB used (cache 0.3, shmem 0.0) | cgroup 4.6/172 GB | disk 144 GB used | gpu 83.1/95.6 GiB, 100%, 288 W | load 0.0 | largest VLLM::EngineCor 4.3 GB, vllm 2.5 GB, python 1.2 GB
 
 
58
  [2026-11-19 04:30:13] ram 0.2/172 GB used (cache 0.0, shmem 0.0) | cgroup 0.2/172 GB | disk 143 GB used | gpu 0.0/95.6 GiB, 0%, 62 W | load 0.0 | largest python 0.1 GB, python 0.1 GB, python 0.0 GB
59
  [2026-11-19 04:30:14] ram 0.2/172 GB used (cache 0.0, shmem 0.0) | cgroup 0.2/172 GB | disk 143 GB used | gpu 0.0/95.6 GiB, 0%, 62 W | load 0.0 | largest python 0.1 GB, python 0.1 GB, python 0.0 GB
60
  [2026-11-19 04:36:46] ram 4.6/172 GB used (cache 0.3, shmem 0.0) | cgroup 4.6/172 GB | disk 144 GB used | gpu 83.1/95.6 GiB, 100%, 288 W | load 0.0 | largest VLLM::EngineCor 4.3 GB, vllm 2.5 GB, python 1.2 GB
61
+ [2026-11-19 04:46:48] ram 4.6/172 GB used (cache 0.3, shmem 0.0) | cgroup 4.6/172 GB | disk 144 GB used | gpu 83.1/95.6 GiB, 100%, 288 W | load 0.0 | largest VLLM::EngineCor 4.3 GB, vllm 2.5 GB, python 1.2 GB
runs/r3/logs/sync.log CHANGED
@@ -47,3 +47,4 @@ uploaded logs/results to mudassiruddin21/EMC:runs/r3
47
  uploaded logs/results to mudassiruddin21/EMC:runs/r3
48
  uploaded logs/results to mudassiruddin21/EMC:runs/r3
49
  uploaded logs/results to mudassiruddin21/EMC:runs/r3
 
 
47
  uploaded logs/results to mudassiruddin21/EMC:runs/r3
48
  uploaded logs/results to mudassiruddin21/EMC:runs/r3
49
  uploaded logs/results to mudassiruddin21/EMC:runs/r3
50
+ uploaded logs/results to mudassiruddin21/EMC:runs/r3