llamaserver-logs-after-rewind.txt

0.00.192.856 I cmn  common_param: common_params_print_info: verbosity = 3 (adjust with the `-lv N` CLI arg)
0.00.427.523 I srv    load_model: loading model 'A:\models\Qwen3.8-27B-NVFP4-MTP-MEDIUM.gguf'
0.09.775.242 I cmn          init: llama threadpool init, n_threads = 6
0.09.934.768 I common_speculative_init_result: creating MTP draft context against the target model 'A:\models\Qwen3.8-27B-NVFP4-MTP-MEDIUM.gguf'
0.13.268.340 I srv    load_model: loaded multimodal model, 'A:\models\Qwen3.8-27B-mmproj-BF16.gguf'
0.13.364.458 I srv    load_model: initializing, n_slots = 1, n_ctx_slot = 180224, kv_unified = 'true'
0.13.676.911 I srv  llama_server: model loaded
0.13.676.918 I srv  llama_server: listening on http://127.0.0.1:8080
0.13.676.919 W srv  llama_server: NOTICE: server default port will be changed to :9931 in a future release
0.13.676.921 W srv  llama_server:         ref: https://github.com/ggml-org/llama.cpp/pull/26508
3.37.780.706 I slot get_availabl: id  0 | task -1 | selected slot by LRU, t_last = -1
3.37.782.926 I slot launch_slot_: id  0 | task 0 | processing task, is_child = 0
3.41.043.761 I slot print_timing: id  0 | task 0 | prompt processing, n_tokens =   8192, progress = 0.30, t =   3.11 s / 2636.87 tokens per second
3.41.791.398 I slot print_timing: id  0 | task 0 | prompt processing, n_tokens =  10240, progress = 0.37, t =   3.84 s / 2668.20 tokens per second
3.42.558.098 I slot print_timing: id  0 | task 0 | prompt processing, n_tokens =  12288, progress = 0.44, t =   4.60 s / 2668.96 tokens per second
3.43.360.502 I slot print_timing: id  0 | task 0 | prompt processing, n_tokens =  14336, progress = 0.52, t =   5.40 s / 2655.25 tokens per second
3.44.164.780 I slot print_timing: id  0 | task 0 | prompt processing, n_tokens =  16384, progress = 0.59, t =   6.20 s / 2641.51 tokens per second
3.45.018.283 I slot print_timing: id  0 | task 0 | prompt processing, n_tokens =  18432, progress = 0.67, t =   7.04 s / 2617.25 tokens per second
3.45.885.918 I slot print_timing: id  0 | task 0 | prompt processing, n_tokens =  20480, progress = 0.74, t =   7.91 s / 2589.73 tokens per second
3.46.793.756 I slot print_timing: id  0 | task 0 | prompt processing, n_tokens =  22528, progress = 0.81, t =   8.80 s / 2559.20 tokens per second
3.47.726.885 I slot print_timing: id  0 | task 0 | prompt processing, n_tokens =  24576, progress = 0.89, t =   9.73 s / 2525.33 tokens per second
3.48.142.576 I slot print_timing: id  0 | task 0 | prompt processing, n_tokens =  25404, progress = 0.92, t =  10.25 s / 2478.64 tokens per second
3.49.075.072 I slot print_timing: id  0 | task 0 | prompt processing, n_tokens =  27153, progress = 0.98, t =  11.12 s / 2440.82 tokens per second
3.49.506.764 I slot print_timing: id  0 | task 0 | prompt processing, n_tokens =  27665, progress = 1.00, t =  11.58 s / 2388.07 tokens per second
3.52.696.514 I slot print_timing: id  0 | task 0 | n_gen =    215, tg =  70.94 t/s, tg_3s =  71.26 t/s
3.55.706.949 I slot print_timing: id  0 | task 0 | n_gen =    368, tg =  60.89 t/s, tg_3s =  50.82 t/s
3.56.334.082 I slot print_timing: id  0 | task 0 | prompt eval time =   11896.62 ms / 27669 tokens (    0.43 ms per token,  2325.79 tokens per second)
3.56.334.088 I slot print_timing: id  0 | task 0 |        eval time =    6654.27 ms /   417 tokens (   16.00 ms per token,    62.52 tokens per second)
3.56.334.089 I slot print_timing: id  0 | task 0 |       total time =   18550.89 ms / 28086 tokens
3.56.334.091 I slot print_timing: id  0 | task 0 |    graphs reused =        130
3.56.334.097 I slot print_timing: id  0 | task 0 | draft acceptance = 0.54771 (  287 accepted /   524 generated), mean len =  3.19
3.56.335.443 I slot      release: id  0 | task 0 | stop processing: n_tokens = 28087, truncated = 0
3.56.710.572 I slot get_availabl: id  0 | task -1 | selected slot by LCP similarity, f_sim_best = 0.993 (> 0.100 thold), f_keep = 1.000
3.56.712.513 I slot launch_slot_: id  0 | task 148 | processing task, is_child = 0
3.59.339.705 I slot print_timing: id  0 | task 148 | prompt eval time =     456.27 ms /   211 tokens (    2.16 ms per token,   462.45 tokens per second)
3.59.339.712 I slot print_timing: id  0 | task 148 |        eval time =    2170.67 ms /   148 tokens (   14.77 ms per token,    67.72 tokens per second)
3.59.339.714 I slot print_timing: id  0 | task 148 |       total time =    2626.93 ms /   359 tokens
3.59.339.715 I slot print_timing: id  0 | task 148 |    graphs reused =        169
3.59.339.722 I slot print_timing: id  0 | task 148 | draft acceptance = 0.65854 (  108 accepted /   164 generated), mean len =  3.63
3.59.341.167 I slot      release: id  0 | task 148 | stop processing: n_tokens = 28447, truncated = 0
3.59.684.914 I slot get_availabl: id  0 | task -1 | selected slot by LCP similarity, f_sim_best = 0.967 (> 0.100 thold), f_keep = 1.000
3.59.687.019 I slot launch_slot_: id  0 | task 192 | processing task, is_child = 0
4.03.594.054 I slot print_timing: id  0 | task 192 | n_gen =    213, tg =  70.33 t/s, tg_3s =  70.66 t/s
4.05.214.335 I slot print_timing: id  0 | task 192 | prompt eval time =     892.52 ms /   957 tokens (    0.93 ms per token,  1072.24 tokens per second)
4.05.214.340 I slot print_timing: id  0 | task 192 |        eval time =    4634.54 ms /   341 tokens (   13.63 ms per token,    73.36 tokens per second)
4.05.214.341 I slot print_timing: id  0 | task 192 |       total time =    5527.07 ms /  1298 tokens
4.05.214.343 I slot print_timing: id  0 | task 192 |    graphs reused =        258
4.05.214.348 I slot print_timing: id  0 | task 192 | draft acceptance = 0.67935 (  250 accepted /   368 generated), mean len =  3.72
4.05.215.352 I slot      release: id  0 | task 192 | stop processing: n_tokens = 29746, truncated = 0
4.05.655.613 I slot get_availabl: id  0 | task -1 | selected slot by LCP similarity, f_sim_best = 0.992 (> 0.100 thold), f_keep = 1.000
4.05.657.836 I slot launch_slot_: id  0 | task 288 | processing task, is_child = 0
4.09.125.868 I slot print_timing: id  0 | task 288 | n_gen =    174, tg =  57.60 t/s, tg_3s =  57.93 t/s
4.11.282.892 I slot print_timing: id  0 | task 288 | prompt eval time =     464.47 ms /   255 tokens (    1.82 ms per token,   549.02 tokens per second)
4.11.282.899 I slot print_timing: id  0 | task 288 |        eval time =    5160.15 ms /   315 tokens (   16.43 ms per token,    60.85 tokens per second)
4.11.282.900 I slot print_timing: id  0 | task 288 |       total time =    5624.61 ms /   570 tokens
4.11.282.901 I slot print_timing: id  0 | task 288 |    graphs reused =        358
4.11.282.908 I slot print_timing: id  0 | task 288 | draft acceptance = 0.52451 (  214 accepted /   408 generated), mean len =  3.10
4.11.284.399 I slot      release: id  0 | task 288 | stop processing: n_tokens = 30317, truncated = 0
4.11.614.956 I slot get_availabl: id  0 | task -1 | selected slot by LCP similarity, f_sim_best = 0.974 (> 0.100 thold), f_keep = 1.000
4.11.617.493 I slot launch_slot_: id  0 | task 393 | processing task, is_child = 0
4.15.509.339 I slot print_timing: id  0 | task 393 | n_gen =    162, tg =  53.51 t/s, tg_3s =  53.84 t/s
4.18.520.271 I slot print_timing: id  0 | task 393 | n_gen =    326, tg =  53.99 t/s, tg_3s =  54.47 t/s
4.21.561.130 I slot print_timing: id  0 | task 393 | n_gen =    496, tg =  54.63 t/s, tg_3s =  55.91 t/s
4.24.594.296 I slot print_timing: id  0 | task 393 | n_gen =    685, tg =  56.56 t/s, tg_3s =  62.31 t/s
4.24.790.919 I slot print_timing: id  0 | task 393 | prompt eval time =     883.01 ms /   819 tokens (    1.08 ms per token,   927.51 tokens per second)
4.24.790.925 I slot print_timing: id  0 | task 393 |        eval time =   12290.17 ms /   705 tokens (   17.46 ms per token,    57.28 tokens per second)
4.24.790.927 I slot print_timing: id  0 | task 393 |       total time =   13173.18 ms /  1524 tokens
4.24.790.928 I slot print_timing: id  0 | task 393 |    graphs reused =        600
4.24.790.934 I slot print_timing: id  0 | task 393 | draft acceptance = 0.46545 (  458 accepted /   984 generated), mean len =  2.86
4.24.792.267 I slot      release: id  0 | task 393 | stop processing: n_tokens = 31840, truncated = 0
4.25.084.601 I slot get_availabl: id  0 | task -1 | selected slot by LCP similarity, f_sim_best = 0.975 (> 0.100 thold), f_keep = 1.000
4.25.086.424 I slot launch_slot_: id  0 | task 643 | processing task, is_child = 0
4.28.988.756 I slot print_timing: id  0 | task 643 | n_gen =    180, tg =  59.25 t/s, tg_3s =  59.57 t/s
4.29.086.255 I slot print_timing: id  0 | task 643 | prompt eval time =     880.85 ms /   826 tokens (    1.07 ms per token,   937.73 tokens per second)
4.29.086.262 I slot print_timing: id  0 | task 643 |        eval time =    3118.58 ms /   190 tokens (   16.50 ms per token,    60.60 tokens per second)
4.29.086.263 I slot print_timing: id  0 | task 643 |       total time =    3999.43 ms /  1016 tokens
4.29.086.265 I slot print_timing: id  0 | task 643 |    graphs reused =        659
4.29.086.271 I slot print_timing: id  0 | task 643 | draft acceptance = 0.52459 (  128 accepted /   244 generated), mean len =  3.10
4.29.088.011 I slot      release: id  0 | task 643 | stop processing: n_tokens = 32855, truncated = 0
4.38.784.413 I slot get_availabl: id  0 | task -1 | selected slot by LCP similarity, f_sim_best = 0.994 (> 0.100 thold), f_keep = 1.000
4.38.786.191 I slot launch_slot_: id  0 | task 708 | processing task, is_child = 0
4.40.499.829 I slot print_timing: id  0 | task 708 | prompt eval time =     793.63 ms /   206 tokens (    3.85 ms per token,   259.57 tokens per second)
4.40.499.836 I slot print_timing: id  0 | task 708 |        eval time =     919.72 ms /    81 tokens (   11.50 ms per token,    86.98 tokens per second)
4.40.499.838 I slot print_timing: id  0 | task 708 |       total time =    1713.35 ms /   287 tokens
4.40.499.839 I slot print_timing: id  0 | task 708 |    graphs reused =        676
4.40.499.846 I slot print_timing: id  0 | task 708 | draft acceptance = 0.87500 (   63 accepted /    72 generated), mean len =  4.50
4.40.501.851 I slot      release: id  0 | task 708 | stop processing: n_tokens = 33142, truncated = 0
4.43.375.068 I slot get_availabl: id  0 | task -1 | selected slot by LCP similarity, f_sim_best = 0.995 (> 0.100 thold), f_keep = 1.000
4.43.377.462 I slot launch_slot_: id  0 | task 729 | processing task, is_child = 0
4.46.909.637 I slot print_timing: id  0 | task 729 | n_gen =    177, tg =  58.20 t/s, tg_3s =  58.53 t/s
4.49.946.313 I slot print_timing: id  0 | task 729 | n_gen =    325, tg =  53.46 t/s, tg_3s =  48.74 t/s
4.51.345.769 I slot print_timing: id  0 | task 729 | prompt eval time =     507.81 ms /   180 tokens (    2.82 ms per token,   354.46 tokens per second)
4.51.345.775 I slot print_timing: id  0 | task 729 |        eval time =    7460.07 ms /   421 tokens (   17.76 ms per token,    56.30 tokens per second)
4.51.345.776 I slot print_timing: id  0 | task 729 |       total time =    7967.89 ms /   601 tokens
4.51.345.777 I slot print_timing: id  0 | task 729 |    graphs reused =        819
4.51.345.784 I slot print_timing: id  0 | task 729 | draft acceptance = 0.47759 (  277 accepted /   580 generated), mean len =  2.91
4.51.347.496 I slot      release: id  0 | task 729 | stop processing: n_tokens = 33744, truncated = 0
4.51.577.868 I slot get_availabl: id  0 | task -1 | selected slot by LCP similarity, f_sim_best = 0.996 (> 0.100 thold), f_keep = 1.000
4.51.579.630 I slot launch_slot_: id  0 | task 877 | processing task, is_child = 0
4.55.007.222 I slot print_timing: id  0 | task 877 | n_gen =    195, tg =  63.63 t/s, tg_3s =  63.95 t/s
4.58.038.746 I slot print_timing: id  0 | task 877 | n_gen =    336, tg =  55.10 t/s, tg_3s =  46.51 t/s
5.01.049.308 I slot print_timing: id  0 | task 877 | n_gen =    531, tg =  58.30 t/s, tg_3s =  64.77 t/s
5.03.713.168 I slot print_timing: id  0 | task 877 | prompt eval time =     378.54 ms /   147 tokens (    2.58 ms per token,   388.33 tokens per second)
5.03.713.173 I slot print_timing: id  0 | task 877 |        eval time =   11754.74 ms /   691 tokens (   17.04 ms per token,    58.70 tokens per second)
5.03.713.173 I slot print_timing: id  0 | task 877 |       total time =   12133.28 ms /   838 tokens
5.03.713.174 I slot print_timing: id  0 | task 877 |    graphs reused =       1047
5.03.713.179 I slot print_timing: id  0 | task 877 | draft acceptance = 0.49569 (  460 accepted /   928 generated), mean len =  2.98
5.03.714.617 I slot      release: id  0 | task 877 | stop processing: n_tokens = 34583, truncated = 0