lmstudio-ai / lmstudio-ai/lmstudio-bug-tracker

v0.4.15-GUI-Gemma 4 26B A4B it compressed not acting as it had been until latest updates

Open
#2,011 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
No language data
Stars
152
Forks
38
PR merge metrics
No merged PRs in 30d

Description

LM Studio 0.4.15
Windows 11
CPU llama.cpp.2.20.1
LLM: Gemma 4 26B A4B it compressed
no plugins/MCPs
Server: Stopped

Prior to updating both LM Studio from 0.4.14 to 0.4.15 and all backends from 2.18.x to 2.20.0 and then to 2.20.1, the response time was about 10 minutes. Same PC, same /preset of hyperparameters, no System Prompt, same-ish Prompt.

Now, after about half an hour (09:42 -> 10:16) of waiting for a reply from the model, looking at the Verbose log I found this gem which I suspect is inside of llama.cpp (at least in its CPU incarnation):

What DnsConfig is trying whoever to read, and to what end exactly?

2026-06-05 09:42:16 [DEBUG]
2.06.645.848 I slot get_availabl: id 0 | task -1 | selected slot by LRU, t_last = -1
2.06.645.859 I srv get_availabl: updating prompt cache
2.06.646.193 I srv load: - looking for better prompt, base f_keep = -1.000, sim = 0.000
2.06.646.628 I srv update: - cache state: 0 prompts, 0.000 MiB (limits: 8192.000 MiB, 60160 tokens, 8589934592 est)
2.06.646.641 I srv get_availabl: prompt cache update took 0.78 ms
2.06.647.516 I slot launch_slot_: id 0 | task 0 | processing task, is_child = 0
2.06.647.564 W slot update_slots: id 0 | task 0 | cache reuse is not supported - ignoring n_cache_reuse = 256
2026-06-05 10:16:22 [DEBUG]
[10488:0605/101622.476:WARNING:net\dns\dns_config_service_win.cc:605] Failed to read DnsConfig.
2026-06-05 10:16:27 [DEBUG]
[10488:0605/101627.497:WARNING:net\dns\dns_config_service_win.cc:605] Failed to read DnsConfig.
2026-06-05 10:16:37 [DEBUG]
[10488:0605/101637.508:WARNING:net\dns\dns_config_service_win.cc:605] Failed to read DnsConfig.
2026-06-05 10:16:57 [DEBUG]
[10488:0605/101657.529:WARNING:net\dns\dns_config_service_win.cc:605] Failed to read DnsConfig.

At 10:27 mark, the reply did come in, eventually.
Bonus, the same flood of IP poking every n tokens - as signaled in https://github.com/lmstudio-ai/lmstudio-bug-tracker/issues/2006

2026-06-05 10:16:22 [DEBUG]
[10488:0605/101622.476:WARNING:net\dns\dns_config_service_win.cc:605] Failed to read DnsConfig.
2026-06-05 10:16:27 [DEBUG]
[10488:0605/101627.497:WARNING:net\dns\dns_config_service_win.cc:605] Failed to read DnsConfig.
2026-06-05 10:16:37 [DEBUG]
[10488:0605/101637.508:WARNING:net\dns\dns_config_service_win.cc:605] Failed to read DnsConfig.
2026-06-05 10:16:57 [DEBUG]
[10488:0605/101657.529:WARNING:net\dns\dns_config_service_win.cc:605] Failed to read DnsConfig.
2026-06-05 10:27:51 [DEBUG]
47.42.244.949 I slot print_timing: id 0 | task 0 | prompt processing, n_tokens = 10020, progress = 1.00, t = 2735.60 s / 3.66 tokens per second
2026-06-05 10:27:51 [DEBUG]
47.42.547.307 I slot print_timing: id 0 | task 0 | prompt processing, n_tokens = 10021, progress = 1.00, t = 2735.90 s / 3.66 tokens per second
2026-06-05 10:27:52 [DEBUG]
47.42.664.133 I slot create_check: id 0 | task 0 | created context checkpoint 1 of 32 (pos_min = 8741, pos_max = 10020, n_tokens = 10021, size = 200.012 MiB)
2026-06-05 10:28:24 [DEBUG]
48.15.547.130 I slot print_timing: id 0 | task 0 | n_decoded = 100, tg = 3.33 t/s
2026-06-05 10:28:28 [DEBUG]
48.18.616.488 I slot print_timing: id 0 | task 0 | n_decoded = 111, tg = 3.36 t/s
2026-06-05 10:28:31 [DEBUG]
48.21.878.721 I slot print_timing: id 0 | task 0 | n_decoded = 122, tg = 3.36 t/s
2026-06-05 10:28:34 [DEBUG]
48.24.994.928 I slot print_timing: id 0 | task 0 | n_decoded = 133, tg = 3.37 t/s
2026-06-05 10:28:37 [DEBUG]
48.28.078.162 I slot print_timing: id 0 | task 0 | n_decoded = 144, tg = 3.39 t/s
2026-06-05 10:28:40 [DEBUG]
48.31.175.931 I slot print_timing: id 0 | task 0 | n_decoded = 155, tg = 3.40 t/s
2026-06-05 10:28:43 [DEBUG]
48.34.249.331 I slot print_timing: id 0 | task 0 | n_decoded = 166, tg = 3.41 t/s
2026-06-05 10:28:46 [DEBUG]
48.37.509.624 I slot print_timing: id 0 | task 0 | n_decoded = 176, tg = 3.39 t/s
2026-06-05 10:28:50 [DEBUG]
48.40.702.308 I slot print_timing: id 0 | task 0 | n_decoded = 187, tg = 3.39 t/s
2026-06-05 10:28:53 [DEBUG]
48.43.906.146 I slot print_timing: id 0 | task 0 | n_decoded = 198, tg = 3.39 t/s
2026-06-05 10:28:56 [DEBUG]
48.47.011.433 I slot print_timing: id 0 | task 0 | n_decoded = 209, tg = 3.40 t/s
2026-06-05 10:28:59 [DEBUG]
48.50.098.679 I slot print_timing: id 0 | task 0 | n_decoded = 220, tg = 3.41 t/s
2026-06-05 10:29:02 [DEBUG]
48.53.214.091 I slot print_timing: id 0 | task 0 | n_decoded = 230, tg = 3.40 t/s
2026-06-05 10:29:05 [DEBUG]
48.56.508.392 I slot print_timing: id 0 | task 0 | n_decoded = 241, tg = 3.40 t/s
2026-06-05 10:29:09 [DEBUG]
48.59.691.048 I slot print_timing: id 0 | task 0 | n_decoded = 252, tg = 3.40 t/s
2026-06-05 10:29:12 [DEBUG]
49.02.846.902 I slot print_timing: id 0 | task 0 | n_decoded = 263, tg = 3.40 t/s
2026-06-05 10:29:15 [DEBUG]
49.05.847.329 I slot print_timing: id 0 | task 0 | n_decoded = 273, tg = 3.40 t/s
2026-06-05 10:29:18 [DEBUG]
49.08.992.961 I slot print_timing: id 0 | task 0 | n_decoded = 284, tg = 3.40 t/s
2026-06-05 10:29:21 [DEBUG]
49.12.046.460 I slot print_timing: id 0 | task 0 | n_decoded = 294, tg = 3.40 t/s
2026-06-05 10:29:24 [DEBUG]
49.15.069.573 I slot print_timing: id 0 | task 0 | n_decoded = 304, tg = 3.40 t/s
2026-06-05 10:29:27 [DEBUG]
49.18.260.333 I slot print_timing: id 0 | task 0 | n_decoded = 315, tg = 3.40 t/s
2026-06-05 10:29:30 [DEBUG]
49.21.291.298 I slot print_timing: id 0 | task 0 | n_decoded = 326, tg = 3.41 t/s
2026-06-05 10:29:33 [DEBUG]
49.24.550.470 I slot print_timing: id 0 | task 0 | n_decoded = 338, tg = 3.41 t/s
2026-06-05 10:29:37 [DEBUG]
49.27.603.259 I slot print_timing: id 0 | task 0 | n_decoded = 349, tg = 3.42 t/s
2026-06-05 10:29:40 [DEBUG]
49.30.680.347 I slot print_timing: id 0 | task 0 | n_decoded = 361, tg = 3.43 t/s
2026-06-05 10:29:43 [DEBUG]
49.33.836.233 I slot print_timing: id 0 | task 0 | n_decoded = 374, tg = 3.45 t/s
2026-06-05 10:29:46 [DEBUG]
49.36.983.154 I slot print_timing: id 0 | task 0 | n_decoded = 387, tg = 3.47 t/s
2026-06-05 10:29:49 [DEBUG]
49.40.046.888 I slot print_timing: id 0 | task 0 | n_decoded = 400, tg = 3.49 t/s
2026-06-05 10:29:52 [DEBUG]
49.43.106.172 I slot print_timing: id 0 | task 0 | n_decoded = 413, tg = 3.51 t/s
2026-06-05 10:29:55 [DEBUG]
49.46.341.469 I slot print_timing: id 0 | task 0 | n_decoded = 426, tg = 3.53 t/s
2026-06-05 10:29:58 [DEBUG]
49.49.490.828 I slot print_timing: id 0 | task 0 | n_decoded = 439, tg = 3.54 t/s
2026-06-05 10:30:02 [DEBUG]
49.52.604.528 I slot print_timing: id 0 | task 0 | n_decoded = 452, tg = 3.56 t/s
2026-06-05 10:30:05 [DEBUG]
49.55.782.978 I slot print_timing: id 0 | task 0 | n_decoded = 465, tg = 3.57 t/s
2026-06-05 10:30:08 [DEBUG]
49.58.800.664 I slot print_timing: id 0 | task 0 | n_decoded = 477, tg = 3.58 t/s
2026-06-05 10:30:11 [DEBUG]
50.01.862.557 I slot print_timing: id 0 | task 0 | n_decoded = 489, tg = 3.59 t/s
2026-06-05 10:30:14 [DEBUG]
50.04.973.336 I slot print_timing: id 0 | task 0 | n_decoded = 501, tg = 3.59 t/s
2026-06-05 10:30:17 [DEBUG]
50.08.213.528 I slot print_timing: id 0 | task 0 | n_decoded = 514, tg = 3.60 t/s
2026-06-05 10:30:20 [DEBUG]
50.11.433.183 I slot print_timing: id 0 | task 0 | n_decoded = 527, tg = 3.61 t/s
2026-06-05 10:30:23 [DEBUG]
50.14.559.590 I slot print_timing: id 0 | task 0 | n_decoded = 540, tg = 3.62 t/s
2026-06-05 10:30:27 [DEBUG]
50.17.793.595 I slot print_timing: id 0 | task 0 | n_decoded = 553, tg = 3.63 t/s
2026-06-05 10:30:30 [DEBUG]
50.20.935.379 I slot print_timing: id 0 | task 0 | n_decoded = 566, tg = 3.64 t/s
2026-06-05 10:30:33 [DEBUG]
50.24.000.676 I slot print_timing: id 0 | task 0 | n_decoded = 579, tg = 3.65 t/s
2026-06-05 10:30:36 [DEBUG]
50.27.223.217 I slot print_timing: id 0 | task 0 | n_decoded = 592, tg = 3.66 t/s
2026-06-05 10:30:39 [DEBUG]
50.30.275.761 I slot print_timing: id 0 | task 0 | n_decoded = 602, tg = 3.65 t/s
2026-06-05 10:30:42 [DEBUG]
50.33.414.878 I slot print_timing: id 0 | task 0 | n_decoded = 611, tg = 3.64 t/s
2026-06-05 10:30:45 [DEBUG]
50.36.483.586 I slot print_timing: id 0 | task 0 | n_decoded = 623, tg = 3.64 t/s
2026-06-05 10:30:48 [DEBUG]
50.39.530.635 I slot print_timing: id 0 | task 0 | n_decoded = 635, tg = 3.65 t/s
2026-06-05 10:30:52 [DEBUG]
50.42.693.801 I slot print_timing: id 0 | task 0 | n_decoded = 648, tg = 3.66 t/s
2026-06-05 10:30:55 [DEBUG]
50.45.727.319 I slot print_timing: id 0 | task 0 | n_decoded = 660, tg = 3.66 t/s
2026-06-05 10:30:58 [DEBUG]
50.48.961.204 I slot print_timing: id 0 | task 0 | n_decoded = 673, tg = 3.67 t/s
2026-06-05 10:31:01 [DEBUG]
50.52.075.446 I slot print_timing: id 0 | task 0 | n_decoded = 686, tg = 3.68 t/s
2026-06-05 10:31:04 [DEBUG]
50.55.251.470 I slot print_timing: id 0 | task 0 | n_decoded = 699, tg = 3.68 t/s
2026-06-05 10:31:07 [DEBUG]
50.58.349.794 I slot print_timing: id 0 | task 0 | n_decoded = 712, tg = 3.69 t/s
2026-06-05 10:31:10 [DEBUG]
51.01.441.137 I slot print_timing: id 0 | task 0 | n_decoded = 725, tg = 3.70 t/s
2026-06-05 10:31:14 [DEBUG]
51.04.634.536 I slot print_timing: id 0 | task 0 | n_decoded = 738, tg = 3.71 t/s
2026-06-05 10:31:17 [DEBUG]
51.07.787.272 I slot print_timing: id 0 | task 0 | n_decoded = 751, tg = 3.71 t/s
2026-06-05 10:31:20 [DEBUG]
51.11.044.431 I slot print_timing: id 0 | task 0 | n_decoded = 764, tg = 3.72 t/s
2026-06-05 10:31:23 [DEBUG]
51.14.093.856 I slot print_timing: id 0 | task 0 | n_decoded = 776, tg = 3.72 t/s
2026-06-05 10:31:26 [DEBUG]
51.17.382.775 I slot print_timing: id 0 | task 0 | n_decoded = 788, tg = 3.72 t/s
2026-06-05 10:31:29 [DEBUG]
51.20.556.970 I slot print_timing: id 0 | task 0 | n_decoded = 798, tg = 3.71 t/s
2026-06-05 10:31:33 [DEBUG]
51.23.796.877 I slot print_timing: id 0 | task 0 | n_decoded = 810, tg = 3.71 t/s
2026-06-05 10:31:36 [DEBUG]
51.27.005.591 I slot print_timing: id 0 | task 0 | n_decoded = 822, tg = 3.71 t/s
2026-06-05 10:31:39 [DEBUG]
51.30.250.431 I slot print_timing: id 0 | task 0 | n_decoded = 835, tg = 3.72 t/s
2026-06-05 10:31:42 [DEBUG]
51.33.441.772 I slot print_timing: id 0 | task 0 | n_decoded = 848, tg = 3.72 t/s
2026-06-05 10:31:46 [DEBUG]
51.36.793.837 I slot print_timing: id 0 | task 0 | n_decoded = 859, tg = 3.71 t/s
2026-06-05 10:31:49 [DEBUG]
51.40.084.421 I slot print_timing: id 0 | task 0 | n_decoded = 868, tg = 3.70 t/s
2026-06-05 10:31:52 [DEBUG]
51.43.347.618 I slot print_timing: id 0 | task 0 | n_decoded = 880, tg = 3.70 t/s
2026-06-05 10:31:55 [DEBUG]
51.46.413.457 I slot print_timing: id 0 | task 0 | n_decoded = 890, tg = 3.70 t/s
2026-06-05 10:31:58 [DEBUG]
51.49.542.673 I slot print_timing: id 0 | task 0 | n_decoded = 902, tg = 3.70 t/s
2026-06-05 10:32:02 [DEBUG]
51.52.649.368 I slot print_timing: id 0 | task 0 | n_decoded = 914, tg = 3.70 t/s
2026-06-05 10:32:05 [DEBUG]
51.55.653.098 I slot print_timing: id 0 | task 0 | n_decoded = 926, tg = 3.70 t/s
2026-06-05 10:32:08 [DEBUG]
51.58.664.898 I slot print_timing: id 0 | task 0 | n_decoded = 938, tg = 3.71 t/s
2026-06-05 10:32:11 [DEBUG]
52.02.005.819 I slot print_timing: id 0 | task 0 | n_decoded = 949, tg = 3.70 t/s
2026-06-05 10:32:14 [DEBUG]
52.05.196.559 I slot print_timing: id 0 | task 0 | n_decoded = 957, tg = 3.69 t/s
2026-06-05 10:32:17 [DEBUG]
52.08.412.717 I slot print_timing: id 0 | task 0 | n_decoded = 969, tg = 3.69 t/s
2026-06-05 10:32:21 [DEBUG]
52.11.591.561 I slot print_timing: id 0 | task 0 | n_decoded = 980, tg = 3.68 t/s
2026-06-05 10:32:24 [DEBUG]
52.14.612.010 I slot print_timing: id 0 | task 0 | n_decoded = 991, tg = 3.68 t/s
2026-06-05 10:32:27 [DEBUG]
52.17.936.094 I slot print_timing: id 0 | task 0 | n_decoded = 1001, tg = 3.68 t/s
2026-06-05 10:32:30 [DEBUG]
52.21.000.533 I slot print_timing: id 0 | task 0 | n_decoded = 1012, tg = 3.67 t/s
2026-06-05 10:32:33 [DEBUG]
52.24.397.600 I slot print_timing: id 0 | task 0 | n_decoded = 1025, tg = 3.68 t/s
2026-06-05 10:32:36 [DEBUG]
52.27.484.353 I slot print_timing: id 0 | task 0 | n_decoded = 1036, tg = 3.67 t/s
2026-06-05 10:32:40 [DEBUG]
52.30.712.416 I slot print_timing: id 0 | task 0 | n_decoded = 1049, tg = 3.68 t/s
2026-06-05 10:32:43 [DEBUG]
52.33.924.765 I slot print_timing: id 0 | task 0 | n_decoded = 1060, tg = 3.68 t/s
2026-06-05 10:32:46 [DEBUG]
52.37.073.049 I slot print_timing: id 0 | task 0 | n_decoded = 1071, tg = 3.67 t/s
2026-06-05 10:32:49 [DEBUG]
52.40.257.205 I slot print_timing: id 0 | task 0 | n_decoded = 1082, tg = 3.67 t/s
2026-06-05 10:32:52 [DEBUG]
52.43.437.700 I slot print_timing: id 0 | task 0 | n_decoded = 1092, tg = 3.67 t/s
2026-06-05 10:32:56 [DEBUG]
52.46.750.555 I slot print_timing: id 0 | task 0 | n_decoded = 1102, tg = 3.66 t/s
2026-06-05 10:32:59 [DEBUG]
52.49.827.569 I slot print_timing: id 0 | task 0 | n_decoded = 1111, tg = 3.65 t/s
2026-06-05 10:33:02 [DEBUG]
52.52.928.368 I slot print_timing: id 0 | task 0 | n_decoded = 1120, tg = 3.64 t/s
2026-06-05 10:33:05 [DEBUG]
52.56.061.034 I slot print_timing: id 0 | task 0 | n_decoded = 1132, tg = 3.65 t/s
2026-06-05 10:33:08 [DEBUG]
52.59.260.605 I slot print_timing: id 0 | task 0 | n_decoded = 1144, tg = 3.65 t/s
2026-06-05 10:33:11 [DEBUG]
53.02.293.846 I slot print_timing: id 0 | task 0 | n_decoded = 1154, tg = 3.64 t/s
2026-06-05 10:33:14 [DEBUG]
53.05.539.688 I slot print_timing: id 0 | task 0 | n_decoded = 1165, tg = 3.64 t/s
2026-06-05 10:33:18 [DEBUG]
53.08.707.264 I slot print_timing: id 0 | task 0 | n_decoded = 1176, tg = 3.64 t/s
2026-06-05 10:33:21 [DEBUG]
53.11.982.160 I slot print_timing: id 0 | task 0 | n_decoded = 1187, tg = 3.64 t/s
2026-06-05 10:33:24 [DEBUG]
53.15.158.327 I slot print_timing: id 0 | task 0 | n_decoded = 1197, tg = 3.63 t/s
2026-06-05 10:33:27 [DEBUG]
53.18.303.239 I slot print_timing: id 0 | task 0 | n_decoded = 1207, tg = 3.63 t/s
2026-06-05 10:33:30 [DEBUG]
53.21.457.718 I slot print_timing: id 0 | task 0 | n_decoded = 1217, tg = 3.62 t/s
2026-06-05 10:33:34 [DEBUG]
53.24.660.442 I slot print_timing: id 0 | task 0 | n_decoded = 1227, tg = 3.62 t/s
2026-06-05 10:33:37 [DEBUG]
53.27.807.885 I slot print_timing: id 0 | task 0 | n_decoded = 1238, tg = 3.62 t/s
2026-06-05 10:33:40 [DEBUG]
53.30.985.035 I slot print_timing: id 0 | task 0 | n_decoded = 1249, tg = 3.62 t/s
2026-06-05 10:33:43 [DEBUG]
53.34.000.215 I slot print_timing: id 0 | task 0 | n_decoded = 1259, tg = 3.61 t/s
2026-06-05 10:33:46 [DEBUG]
53.37.188.841 I slot print_timing: id 0 | task 0 | n_decoded = 1269, tg = 3.61 t/s
2026-06-05 10:33:49 [DEBUG]
53.40.342.401 I slot print_timing: id 0 | task 0 | n_decoded = 1279, tg = 3.61 t/s
2026-06-05 10:33:53 [DEBUG]
53.43.644.593 I slot print_timing: id 0 | task 0 | n_decoded = 1289, tg = 3.60 t/s
2026-06-05 10:33:56 [DEBUG]
53.46.689.663 I slot print_timing: id 0 | task 0 | n_decoded = 1299, tg = 3.60 t/s
2026-06-05 10:33:59 [DEBUG]
53.49.845.559 I slot print_timing: id 0 | task 0 | n_decoded = 1310, tg = 3.60 t/s
2026-06-05 10:34:02 [DEBUG]
53.52.964.293 I slot print_timing: id 0 | task 0 | n_decoded = 1322, tg = 3.60 t/s
2026-06-05 10:34:05 [DEBUG]
53.56.068.202 I slot print_timing: id 0 | task 0 | n_decoded = 1334, tg = 3.60 t/s
2026-06-05 10:34:08 [DEBUG]
53.59.146.580 I slot print_timing: id 0 | task 0 | n_decoded = 1346, tg = 3.60 t/s
2026-06-05 10:34:11 [DEBUG]
54.02.299.154 I slot print_timing: id 0 | task 0 | n_decoded = 1358, tg = 3.60 t/s
2026-06-05 10:34:14 [DEBUG]
54.05.399.248 I slot print_timing: id 0 | task 0 | n_decoded = 1369, tg = 3.60 t/s
2026-06-05 10:34:18 [DEBUG]
54.08.658.655 I slot print_timing: id 0 | task 0 | n_decoded = 1382, tg = 3.61 t/s
2026-06-05 10:34:21 [DEBUG]
54.11.658.929 I slot print_timing: id 0 | task 0 | n_decoded = 1394, tg = 3.61 t/s
2026-06-05 10:34:24 [DEBUG]
54.14.691.330 I slot print_timing: id 0 | task 0 | n_decoded = 1405, tg = 3.61 t/s
2026-06-05 10:34:27 [DEBUG]
54.17.895.762 I slot print_timing: id 0 | task 0 | n_decoded = 1417, tg = 3.61 t/s
2026-06-05 10:34:30 [DEBUG]
54.20.905.366 I slot print_timing: id 0 | task 0 | n_decoded = 1427, tg = 3.61 t/s
2026-06-05 10:34:33 [DEBUG]
54.24.060.999 I slot print_timing: id 0 | task 0 | n_decoded = 1437, tg = 3.61 t/s
2026-06-05 10:34:36 [DEBUG]
54.27.142.167 I slot print_timing: id 0 | task 0 | n_decoded = 1448, tg = 3.61 t/s
2026-06-05 10:34:39 [DEBUG]
54.30.374.335 I slot print_timing: id 0 | task 0 | n_decoded = 1459, tg = 3.60 t/s
2026-06-05 10:34:42 [DEBUG]
54.33.557.914 I slot print_timing: id 0 | task 0 | n_decoded = 1470, tg = 3.60 t/s
2026-06-05 10:34:46 [DEBUG]
54.36.710.266 I slot print_timing: id 0 | task 0 | n_decoded = 1481, tg = 3.60 t/s
2026-06-05 10:34:49 [DEBUG]
54.39.786.279 I slot print_timing: id 0 | task 0 | n_decoded = 1493, tg = 3.60 t/s
2026-06-05 10:34:52 [DEBUG]
54.42.838.313 I slot print_timing: id 0 | task 0 | n_decoded = 1505, tg = 3.61 t/s
2026-06-05 10:34:55 [DEBUG]
54.45.973.272 I slot print_timing: id 0 | task 0 | n_decoded = 1517, tg = 3.61 t/s
2026-06-05 10:34:58 [DEBUG]
54.49.223.626 I slot print_timing: id 0 | task 0 | n_decoded = 1529, tg = 3.61 t/s
2026-06-05 10:35:01 [DEBUG]
54.52.416.305 I slot print_timing: id 0 | task 0 | n_decoded = 1541, tg = 3.61 t/s
2026-06-05 10:35:04 [DEBUG]
54.55.425.842 I slot print_timing: id 0 | task 0 | n_decoded = 1551, tg = 3.61 t/s
2026-06-05 10:35:08 [DEBUG]
54.58.717.972 I slot print_timing: id 0 | task 0 | n_decoded = 1559, tg = 3.60 t/s
2026-06-05 10:35:11 [DEBUG]
55.01.868.454 I slot print_timing: id 0 | task 0 | n_decoded = 1569, tg = 3.60 t/s
2026-06-05 10:35:14 [DEBUG]
55.05.282.911 I slot print_timing: id 0 | task 0 | n_decoded = 1575, tg = 3.58 t/s
2026-06-05 10:35:17 [DEBUG]
55.08.347.964 I slot print_timing: id 0 | task 0 | n_decoded = 1581, tg = 3.57 t/s
2026-06-05 10:35:20 [DEBUG]
55.11.401.160 I slot print_timing: id 0 | task 0 | n_decoded = 1588, tg = 3.56 t/s
2026-06-05 10:35:24 [DEBUG]
55.14.779.377 I slot print_timing: id 0 | task 0 | n_decoded = 1598, tg = 3.56 t/s
2026-06-05 10:35:27 [DEBUG]
55.17.890.246 I slot print_timing: id 0 | task 0 | n_decoded = 1607, tg = 3.55 t/s
...
...
[continued a la long]
...
...
2026-06-05 10:35:51 [DEBUG]
55.42.280.480 I slot print_timing: id 0 | task 0 | prompt eval time = 2738913.08 ms / 10025 tokens ( 273.21 ms per token, 3.66 tokens per second)
55.42.280.493 I slot print_timing: id 0 | task 0 | eval time = 476719.83 ms / 1696 tokens ( 281.08 ms per token, 3.56 tokens per second)
55.42.280.495 I slot print_timing: id 0 | task 0 | total time = 3215632.90 ms / 11721 tokens
55.42.280.496 I slot print_timing: id 0 | task 0 | graphs reused = 11673
2026-06-05 10:35:51 [DEBUG]
55.42.320.803 I slot release: id 0 | task 0 | stop processing: n_tokens = 11720, truncated = 0
2026-06-05 10:35:51 [DEBUG]
55.42.321.539 I srv update_slots: all slots are idle
2026-06-05 10:35:51 [DEBUG]
LlamaV4: server assigned slot
2026-06-05 10:35:51 [DEBUG]
0 to task 0
2026-06-05 10:35:51 [DEBUG]

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start with the supplied verbose log and reproduce the delay on Windows 11 using LM Studio 0.4.15, the CPU backend, and the Gemma 4 26B A4B it compressed model. Compare behavior with LM Studio 0.4.14 and the earlier backend versions, including the DNS warnings and the related issue #2006. Done means the regression is isolated and the model responds without the reported multi-minute delay.

Written by the indexing model from the issue text.

Assessment

Domain
desktop
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Quiet
Clarity
Needs clarification
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.