Run r3: logs and results
Browse files- runs/r3/engine_logs/20261119-033339-eval-r1.log +339 -0
- runs/r3/logs/eval_main.log +3 -0
- runs/r3/logs/stats.log +1 -0
- runs/r3/logs/sync.log +1 -0
runs/r3/engine_logs/20261119-033339-eval-r1.log
CHANGED
|
@@ -93,3 +93,342 @@ Loading safetensors checkpoint shards: 100% Completed | 1/1 [00:00<00:00, 4.48i
|
|
| 93 |
(EngineCore pid=36713) INFO 11-19 03:36:30 [backends.py:916] collected artifacts: 65 entries, 20 artifacts, 24631805 bytes total
|
| 94 |
(EngineCore pid=36713) INFO 11-19 03:36:30 [decorators.py:717] saved AOT compiled function to /workspace/emc/work/cache/vllm/torch_compile_cache/torch_aot_compile/2bf75a0f005d511d44b2e0ebedfdffd0cbe6a7f8d0aa3da736fdfd02e918092c/rank_0_0/model
|
| 95 |
(EngineCore pid=36713) INFO 11-19 03:36:30 [monitor.py:53] torch.compile took 24.47 s in total
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 93 |
(EngineCore pid=36713) INFO 11-19 03:36:30 [backends.py:916] collected artifacts: 65 entries, 20 artifacts, 24631805 bytes total
|
| 94 |
(EngineCore pid=36713) INFO 11-19 03:36:30 [decorators.py:717] saved AOT compiled function to /workspace/emc/work/cache/vllm/torch_compile_cache/torch_aot_compile/2bf75a0f005d511d44b2e0ebedfdffd0cbe6a7f8d0aa3da736fdfd02e918092c/rank_0_0/model
|
| 95 |
(EngineCore pid=36713) INFO 11-19 03:36:30 [monitor.py:53] torch.compile took 24.47 s in total
|
| 96 |
+
(EngineCore pid=36713) INFO 11-19 03:36:44 [monitor.py:81] Initial profiling/warmup run took 13.54 s
|
| 97 |
+
(EngineCore pid=36713) INFO 11-19 03:36:44 [backends.py:1090] Using cache directory: /workspace/emc/work/cache/vllm/torch_compile_cache/cb0ac8f950/rank_0_0/eagle_head for vLLM's torch.compile
|
| 98 |
+
(EngineCore pid=36713) INFO 11-19 03:36:44 [backends.py:1151] Dynamo bytecode transform time: 0.37 s
|
| 99 |
+
(EngineCore pid=36713) INFO 11-19 03:36:46 [backends.py:391] Compiling a graph for compile range (1, 8192) takes 1.85 s
|
| 100 |
+
(EngineCore pid=36713) INFO 11-19 03:36:46 [backends.py:916] collected artifacts: 2 entries, 2 artifacts, 3661471 bytes total
|
| 101 |
+
(EngineCore pid=36713) INFO 11-19 03:36:46 [decorators.py:717] saved AOT compiled function to /workspace/emc/work/cache/vllm/torch_compile_cache/torch_aot_compile/18479e3edf45fa305490e52c44dc54aa80cc7798504d276363b2fcb0c5cc2807/rank_0_0/model
|
| 102 |
+
(EngineCore pid=36713) INFO 11-19 03:36:46 [monitor.py:53] torch.compile took 2.41 s in total
|
| 103 |
+
(EngineCore pid=36713) INFO 11-19 03:36:46 [monitor.py:81] Initial profiling/warmup run took 0.09 s
|
| 104 |
+
(EngineCore pid=36713) WARNING 11-19 03:36:47 [kv_cache_utils.py:1574] Add 3 padding layers, may waste at most 6.25% KV cache memory
|
| 105 |
+
|
| 106 |
+
Capturing CUDA graphs (PIECEWISE): 0%| | 0/7 [00:00<?, ?it/s]
|
| 107 |
+
Capturing CUDA graphs (PIECEWISE): 14%|█▍ | 1/7 [00:06<00:36, 6.14s/it]
|
| 108 |
+
Capturing CUDA graphs (PIECEWISE): 29%|██▊ | 2/7 [00:06<00:13, 2.61s/it]
|
| 109 |
+
Capturing CUDA graphs (PIECEWISE): 43%|████▎ | 3/7 [00:06<00:05, 1.48s/it]
|
| 110 |
+
Capturing CUDA graphs (PIECEWISE): 57%|█████▋ | 4/7 [00:06<00:02, 1.05it/s]
|
| 111 |
+
Capturing CUDA graphs (PIECEWISE): 71%|███████▏ | 5/7 [00:06<00:01, 1.44it/s]
|
| 112 |
+
Capturing CUDA graphs (PIECEWISE): 100%|██████████| 7/7 [00:07<00:00, 2.55it/s]
|
| 113 |
+
Capturing CUDA graphs (PIECEWISE): 100%|██████████| 7/7 [00:07<00:00, 1.00s/it]
|
| 114 |
+
|
| 115 |
+
Capturing CUDA graphs (FULL): 0%| | 0/2 [00:00<?, ?it/s]
|
| 116 |
+
Capturing CUDA graphs (FULL): 100%|██████████| 2/2 [00:00<00:00, 13.83it/s]
|
| 117 |
+
Capturing CUDA graphs (FULL): 100%|██████████| 2/2 [00:00<00:00, 13.80it/s]
|
| 118 |
+
(EngineCore pid=36713) INFO 11-19 03:36:55 [speculator.py:144] Capturing model for speculator...
|
| 119 |
+
|
| 120 |
+
Capturing prefill CUDA graphs (PIECEWISE): 0%| | 0/7 [00:00<?, ?it/s]
|
| 121 |
+
Capturing prefill CUDA graphs (PIECEWISE): 100%|██████████| 7/7 [00:00<00:00, 151.34it/s]
|
| 122 |
+
|
| 123 |
+
Capturing prefill CUDA graphs (FULL): 0%| | 0/3 [00:00<?, ?it/s]
|
| 124 |
+
Capturing prefill CUDA graphs (FULL): 100%|██████████| 3/3 [00:00<00:00, 172.77it/s]
|
| 125 |
+
|
| 126 |
+
Capturing decode CUDA graphs (FULL): 0%| | 0/3 [00:00<?, ?it/s]
|
| 127 |
+
Capturing decode CUDA graphs (FULL): 100%|██████████| 3/3 [00:00<00:00, 95.48it/s]
|
| 128 |
+
(EngineCore pid=36713) INFO 11-19 03:36:55 [model_runner.py:1087] Graph capturing finished in 8 secs, took 0.23 GiB
|
| 129 |
+
(EngineCore pid=36713) INFO 11-19 03:36:56 [gpu_worker.py:692] Available KV cache memory: 30.91 GiB
|
| 130 |
+
(EngineCore pid=36713) INFO 11-19 03:36:56 [gpu_worker.py:707] CUDA graph memory profiling is enabled (default since v0.21.0). The current --gpu-memory-utilization=0.9000 is equivalent to --gpu-memory-utilization=0.8975 without CUDA graph memory profiling. To maintain the same effective KV cache size as before, increase --gpu-memory-utilization to 0.9025. To disable, set VLLM_MEMORY_PROFILER_ESTIMATE_CUDAGRAPHS=0.
|
| 131 |
+
(EngineCore pid=36713) WARNING 11-19 03:36:56 [kv_cache_utils.py:1574] Add 3 padding layers, may waste at most 6.25% KV cache memory
|
| 132 |
+
(EngineCore pid=36713) INFO 11-19 03:36:56 [kv_cache_utils.py:2464] GPU KV cache size: 295,408 tokens, Maximum concurrency for 16,384 tokens per request: 18.03x
|
| 133 |
+
(EngineCore pid=36713) INFO 11-19 03:36:56 [kernel_warmup.py:172] JIT kernel warmup starting.
|
| 134 |
+
(EngineCore pid=36713) INFO 11-19 03:36:56 [kernel_warmup.py:185] JIT kernel warmup finished in 0.00s.
|
| 135 |
+
(EngineCore pid=36713) INFO 11-19 03:36:56 [qwen_triton_warmup.py:294] Warming up Qwen GDN Triton kernels for model_type=qwen3_5_text.
|
| 136 |
+
(EngineCore pid=36713) INFO 11-19 03:36:56 [qwen_vl_triton_warmup.py:98] Warmed M-RoPE Triton kernels.
|
| 137 |
+
(EngineCore pid=36713) INFO 11-19 03:36:56 [mamba_triton_warmup.py:42] Warmed Mamba batch_memcpy_kernel.
|
| 138 |
+
(EngineCore pid=36713) INFO 11-19 03:36:57 [kernel_warmup.py:437] Using FlashInfer autotune cache file: /workspace/emc/work/cache/vllm/flashinfer_autotune_cache/0.7.0.post1/120f/fd30e073b649e1eb00be702764c5abb39723166a735906b2f854d86e1534bb12/autotune_configs.json
|
| 139 |
+
2026-11-19 03:36:57,560 - INFO - autotuner.py:1046 - flashinfer.jit: [Autotuner]: Autotuning process starts ...
|
| 140 |
+
(EngineCore pid=36713) INFO 11-19 03:36:57 [kernel_warmup.py:359] Running FlashInfer autotune with 8192 tokens and token buckets (1, 2, 4, 8, 16, 32, 64, 128, 256, 512, 768, 1024, 1280, 1536, 1792, 2048, 2560, 3072, 3584, 4096, 8192).
|
| 141 |
+
2026-11-19 03:36:57,586 - INFO - autotuner.py:1069 - flashinfer.jit: [Autotuner]: Autotuning process ends
|
| 142 |
+
2026-11-19 03:36:57,740 - INFO - autotuner.py:3554 - flashinfer.jit: [Autotuner]: Saved 0 configs to /workspace/emc/work/cache/vllm/flashinfer_autotune_cache/0.7.0.post1/120f/fd30e073b649e1eb00be702764c5abb39723166a735906b2f854d86e1534bb12/autotune_configs.json (0 new, 0 from previous config)
|
| 143 |
+
|
| 144 |
+
Capturing CUDA graphs (PIECEWISE): 0%| | 0/7 [00:00<?, ?it/s]
|
| 145 |
+
Capturing CUDA graphs (PIECEWISE): 14%|█▍ | 1/7 [00:00<00:00, 6.45it/s]
|
| 146 |
+
Capturing CUDA graphs (PIECEWISE): 29%|██▊ | 2/7 [00:00<00:00, 6.80it/s]
|
| 147 |
+
Capturing CUDA graphs (PIECEWISE): 43%|████▎ | 3/7 [00:00<00:00, 6.92it/s]
|
| 148 |
+
Capturing CUDA graphs (PIECEWISE): 57%|█████▋ | 4/7 [00:00<00:00, 7.05it/s]
|
| 149 |
+
Capturing CUDA graphs (PIECEWISE): 86%|████████▌ | 6/7 [00:00<00:00, 8.59it/s]
|
| 150 |
+
Capturing CUDA graphs (PIECEWISE): 100%|██████████| 7/7 [00:00<00:00, 8.23it/s]
|
| 151 |
+
|
| 152 |
+
Capturing CUDA graphs (FULL): 0%| | 0/3 [00:00<?, ?it/s]
|
| 153 |
+
Capturing CUDA graphs (FULL): 67%|██████▋ | 2/3 [00:00<00:00, 15.91it/s]
|
| 154 |
+
Capturing CUDA graphs (FULL): 100%|██████████| 3/3 [00:00<00:00, 15.71it/s]
|
| 155 |
+
(EngineCore pid=36713) INFO 11-19 03:37:00 [speculator.py:144] Capturing model for speculator...
|
| 156 |
+
|
| 157 |
+
Capturing prefill CUDA graphs (PIECEWISE): 0%| | 0/7 [00:00<?, ?it/s]
|
| 158 |
+
Capturing prefill CUDA graphs (PIECEWISE): 100%|██████████| 7/7 [00:00<00:00, 150.36it/s]
|
| 159 |
+
|
| 160 |
+
Capturing prefill CUDA graphs (FULL): 0%| | 0/3 [00:00<?, ?it/s]
|
| 161 |
+
Capturing prefill CUDA graphs (FULL): 100%|██████████| 3/3 [00:00<00:00, 185.82it/s]
|
| 162 |
+
|
| 163 |
+
Capturing decode CUDA graphs (FULL): 0%| | 0/3 [00:00<?, ?it/s]
|
| 164 |
+
Capturing decode CUDA graphs (FULL): 100%|██████████| 3/3 [00:00<00:00, 111.80it/s]
|
| 165 |
+
(EngineCore pid=36713) INFO 11-19 03:37:01 [model_runner.py:1087] Graph capturing finished in 2 secs, took 0.17 GiB
|
| 166 |
+
(EngineCore pid=36713) INFO 11-19 03:37:01 [gpu_worker.py:892] CUDA graph pool memory: 0.17 GiB (actual), 0.24 GiB (estimated), difference: 0.07 GiB (40.7%).
|
| 167 |
+
(EngineCore pid=36713) INFO 11-19 03:37:01 [gpu_worker.py:955] Free memory on device (94.43/94.97 GiB) on startup. Desired GPU memory utilization is (0.9, 85.48 GiB). Actual usage is 51.37 GiB for consumed memory (weights + non-torch), 3.2 GiB for peak activation, and 0.17 GiB for CUDAGraph memory. Replace gpu_memory_utilization config with `--kv-cache-memory=32850334516` (30.59 GiB) to fit into requested memory, or `--kv-cache-memory=42462617600` (39.55 GiB) to fully utilize gpu memory. Current kv cache memory in use is 30.91 GiB.
|
| 168 |
+
(EngineCore pid=36713) INFO 11-19 03:37:02 [jit_monitor.py:84] Kernel JIT monitor activated; monitored JIT compilations during inference will use mode=warn.
|
| 169 |
+
(EngineCore pid=36713) INFO 11-19 03:37:03 [core.py:373] init engine (profile, create kv cache, warmup model) took 58.85 s (compilation: 26.88 s)
|
| 170 |
+
(EngineCore pid=36713) INFO 11-19 03:37:03 [kv_cache_utils.py:772] kv cache group sizes [16384, 16384, 16384, 800]
|
| 171 |
+
(EngineCore pid=36713) INFO 11-19 03:37:03 [kv_cache_utils.py:773] kv lcm block sizes 409600
|
| 172 |
+
(EngineCore pid=36713) INFO 11-19 03:37:03 [kernel.py:408] Final IR op priority after setting platform defaults: IrOpPriorityConfig(rms_norm=['native'], fused_add_rms_norm=['native'], gelu_and_mul_sparse=['triton', 'native'])
|
| 173 |
+
(APIServer pid=36395) INFO 11-19 03:37:03 [entry.py:132] Supported tasks: ['generate']
|
| 174 |
+
(APIServer pid=36395) INFO 11-19 03:37:03 [factories.py:76] Scale-out endpoints are disabled. Set --enable-scale-out to enable them.
|
| 175 |
+
(APIServer pid=36395) INFO 11-19 03:37:04 [hf.py:646] Detected the chat template content format to be 'openai'. You can set `--chat-template-content-format` to override this.
|
| 176 |
+
(APIServer pid=36395) WARNING 11-19 03:37:04 [serving.py:221] The chat template requires system-first ordering, so inline system messages in /v1/messages requests (e.g. host Code's per-turn reminders) are merged into the leading system prompt. Each new one changes the prompt prefix, so the rest of the conversation misses the prefix cache. Pass a --chat-template that accepts non-leading system messages to avoid this.
|
| 177 |
+
(APIServer pid=36395) WARNING 11-19 03:37:04 [serving.py:221] The chat template requires system-first ordering, so inline system messages in /v1/messages requests (e.g. host Code's per-turn reminders) are merged into the leading system prompt. Each new one changes the prompt prefix, so the rest of the conversation misses the prefix cache. Pass a --chat-template that accepts non-leading system messages to avoid this.
|
| 178 |
+
(APIServer pid=36395) INFO 11-19 03:37:04 [entry.py:136] Starting vLLM server on http://127.0.0.1:35650
|
| 179 |
+
(APIServer pid=36395) INFO 11-19 03:37:04 [launcher.py:60] Available routes are:
|
| 180 |
+
(APIServer pid=36395) INFO 11-19 03:37:04 [launcher.py:69] Route: /openapi.json, Methods: HEAD, GET
|
| 181 |
+
(APIServer pid=36395) INFO 11-19 03:37:04 [launcher.py:69] Route: /docs, Methods: HEAD, GET
|
| 182 |
+
(APIServer pid=36395) INFO 11-19 03:37:04 [launcher.py:69] Route: /docs/oauth2-redirect, Methods: HEAD, GET
|
| 183 |
+
(APIServer pid=36395) INFO 11-19 03:37:04 [launcher.py:69] Route: /redoc, Methods: HEAD, GET
|
| 184 |
+
(APIServer pid=36395) INFO 11-19 03:37:04 [launcher.py:69] Route: /load, Methods: GET
|
| 185 |
+
(APIServer pid=36395) INFO 11-19 03:37:04 [launcher.py:69] Route: /version, Methods: GET
|
| 186 |
+
(APIServer pid=36395) INFO 11-19 03:37:04 [launcher.py:69] Route: /kv_event_sources, Methods: GET
|
| 187 |
+
(APIServer pid=36395) INFO 11-19 03:37:04 [launcher.py:69] Route: /health, Methods: GET
|
| 188 |
+
(APIServer pid=36395) INFO 11-19 03:37:04 [launcher.py:69] Route: /metrics, Methods: GET
|
| 189 |
+
(APIServer pid=36395) INFO 11-19 03:37:04 [launcher.py:69] Route: /tokenize, Methods: POST
|
| 190 |
+
(APIServer pid=36395) INFO 11-19 03:37:04 [launcher.py:69] Route: /detokenize, Methods: POST
|
| 191 |
+
(APIServer pid=36395) INFO 11-19 03:37:04 [launcher.py:69] Route: /v1/models, Methods: GET
|
| 192 |
+
(APIServer pid=36395) INFO 11-19 03:37:04 [launcher.py:69] Route: /ping, Methods: GET
|
| 193 |
+
(APIServer pid=36395) INFO 11-19 03:37:04 [launcher.py:69] Route: /ping, Methods: POST
|
| 194 |
+
(APIServer pid=36395) INFO 11-19 03:37:04 [launcher.py:69] Route: /invocations, Methods: POST
|
| 195 |
+
(APIServer pid=36395) INFO 11-19 03:37:04 [launcher.py:69] Route: /v1/chat/completions, Methods: POST
|
| 196 |
+
(APIServer pid=36395) INFO 11-19 03:37:04 [launcher.py:69] Route: /v1/chat/completions/batch, Methods: POST
|
| 197 |
+
(APIServer pid=36395) INFO 11-19 03:37:04 [launcher.py:69] Route: /v1/responses, Methods: POST
|
| 198 |
+
(APIServer pid=36395) INFO 11-19 03:37:04 [launcher.py:69] Route: /v1/responses/{response_id}, Methods: GET
|
| 199 |
+
(APIServer pid=36395) INFO 11-19 03:37:04 [launcher.py:69] Route: /v1/responses/{response_id}/cancel, Methods: POST
|
| 200 |
+
(APIServer pid=36395) INFO 11-19 03:37:04 [launcher.py:69] Route: /v1/completions, Methods: POST
|
| 201 |
+
(APIServer pid=36395) INFO 11-19 03:37:04 [launcher.py:69] Route: /v1/messages, Methods: POST
|
| 202 |
+
(APIServer pid=36395) INFO 11-19 03:37:04 [launcher.py:69] Route: /v1/messages/count_tokens, Methods: POST
|
| 203 |
+
(APIServer pid=36395) INFO 11-19 03:37:04 [launcher.py:69] Route: /generative_scoring, Methods: POST
|
| 204 |
+
(APIServer pid=36395) INFO 11-19 03:37:04 [launcher.py:69] Route: /scale_elastic_ep, Methods: POST
|
| 205 |
+
(APIServer pid=36395) INFO 11-19 03:37:04 [launcher.py:69] Route: /is_scaling_elastic_ep, Methods: POST
|
| 206 |
+
INFO: Started server process [36395]
|
| 207 |
+
INFO: Waiting for application startup.
|
| 208 |
+
INFO: Application startup complete.
|
| 209 |
+
INFO: 127.0.0.1:62659 - "GET /health HTTP/1.1" 200 OK
|
| 210 |
+
INFO: 127.0.0.1:54999 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 211 |
+
(EngineCore pid=36713) WARNING 11-19 03:37:05 [jit_monitor.py:140] Triton kernel JIT compilation during inference: _compute_local_logits_stats_kernel. This causes a latency spike; consider extending warmup to cover this shape/config.
|
| 212 |
+
(EngineCore pid=36713) WARNING 11-19 03:37:05 [jit_monitor.py:140] Triton kernel JIT compilation during inference: _rejection_kernel. This causes a latency spike; consider extending warmup to cover this shape/config.
|
| 213 |
+
(EngineCore pid=36713) WARNING 11-19 03:37:05 [jit_monitor.py:140] Triton kernel JIT compilation during inference: _resample_kernel. This causes a latency spike; consider extending warmup to cover this shape/config.
|
| 214 |
+
INFO: 127.0.0.1:56441 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 215 |
+
INFO: 127.0.0.1:26581 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 216 |
+
(APIServer pid=36395) INFO 11-19 03:37:14 [loggers.py:320] Engine 000: Avg prompt throughput: 29.2 tokens/s, Avg generation throughput: 47.2 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 217 |
+
(APIServer pid=36395) INFO 11-19 03:37:14 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.29, Accepted throughput: 32.84 tokens/s, Drafted throughput: 43.02 tokens/s, Accepted: 371 tokens, Drafted: 486 tokens, Per-position acceptance rate: 0.914, 0.753, 0.623, Avg Draft acceptance rate: 76.3%
|
| 218 |
+
INFO: 127.0.0.1:54765 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 219 |
+
INFO: 127.0.0.1:22622 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 220 |
+
INFO: 127.0.0.1:37246 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 221 |
+
(APIServer pid=36395) INFO 11-19 03:37:24 [loggers.py:320] Engine 000: Avg prompt throughput: 38.4 tokens/s, Avg generation throughput: 67.6 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 222 |
+
(APIServer pid=36395) INFO 11-19 03:37:24 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.79, Accepted throughput: 49.89 tokens/s, Drafted throughput: 53.69 tokens/s, Accepted: 499 tokens, Drafted: 537 tokens, Per-position acceptance rate: 0.989, 0.933, 0.866, Avg Draft acceptance rate: 92.9%
|
| 223 |
+
INFO: 127.0.0.1:43810 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 224 |
+
(APIServer pid=36395) INFO 11-19 03:37:34 [loggers.py:320] Engine 000: Avg prompt throughput: 3.9 tokens/s, Avg generation throughput: 50.0 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 225 |
+
(APIServer pid=36395) INFO 11-19 03:37:34 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.74, Accepted throughput: 31.70 tokens/s, Drafted throughput: 54.59 tokens/s, Accepted: 317 tokens, Drafted: 546 tokens, Per-position acceptance rate: 0.769, 0.582, 0.390, Avg Draft acceptance rate: 58.1%
|
| 226 |
+
(APIServer pid=36395) INFO 11-19 03:37:44 [loggers.py:320] Engine 000: Avg prompt throughput: 0.0 tokens/s, Avg generation throughput: 48.4 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.4%
|
| 227 |
+
(APIServer pid=36395) INFO 11-19 03:37:44 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.63, Accepted throughput: 30.00 tokens/s, Drafted throughput: 55.19 tokens/s, Accepted: 300 tokens, Drafted: 552 tokens, Per-position acceptance rate: 0.772, 0.505, 0.353, Avg Draft acceptance rate: 54.3%
|
| 228 |
+
INFO: 127.0.0.1:41963 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 229 |
+
(APIServer pid=36395) INFO 11-19 03:37:54 [loggers.py:320] Engine 000: Avg prompt throughput: 7.8 tokens/s, Avg generation throughput: 45.8 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 230 |
+
(APIServer pid=36395) INFO 11-19 03:37:54 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.50, Accepted throughput: 27.60 tokens/s, Drafted throughput: 55.19 tokens/s, Accepted: 276 tokens, Drafted: 552 tokens, Per-position acceptance rate: 0.755, 0.489, 0.255, Avg Draft acceptance rate: 50.0%
|
| 231 |
+
(APIServer pid=36395) INFO 11-19 03:38:04 [loggers.py:320] Engine 000: Avg prompt throughput: 0.0 tokens/s, Avg generation throughput: 47.7 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.4%
|
| 232 |
+
(APIServer pid=36395) INFO 11-19 03:38:04 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.59, Accepted throughput: 29.30 tokens/s, Drafted throughput: 55.20 tokens/s, Accepted: 293 tokens, Drafted: 552 tokens, Per-position acceptance rate: 0.783, 0.527, 0.283, Avg Draft acceptance rate: 53.1%
|
| 233 |
+
INFO: 127.0.0.1:22473 - "GET /metrics HTTP/1.1" 200 OK
|
| 234 |
+
INFO: 127.0.0.1:22398 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 235 |
+
(APIServer pid=36395) INFO 11-19 03:38:14 [loggers.py:320] Engine 000: Avg prompt throughput: 7.2 tokens/s, Avg generation throughput: 47.7 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 236 |
+
(APIServer pid=36395) INFO 11-19 03:38:14 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.62, Accepted throughput: 29.40 tokens/s, Drafted throughput: 54.60 tokens/s, Accepted: 294 tokens, Drafted: 546 tokens, Per-position acceptance rate: 0.813, 0.489, 0.313, Avg Draft acceptance rate: 53.8%
|
| 237 |
+
INFO: 127.0.0.1:18843 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 238 |
+
(APIServer pid=36395) INFO 11-19 03:38:24 [loggers.py:320] Engine 000: Avg prompt throughput: 3.9 tokens/s, Avg generation throughput: 56.2 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 239 |
+
(APIServer pid=36395) INFO 11-19 03:38:24 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.04, Accepted throughput: 37.80 tokens/s, Drafted throughput: 55.49 tokens/s, Accepted: 378 tokens, Drafted: 555 tokens, Per-position acceptance rate: 0.827, 0.676, 0.541, Avg Draft acceptance rate: 68.1%
|
| 240 |
+
INFO: 127.0.0.1:42368 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 241 |
+
(APIServer pid=36395) INFO 11-19 03:38:34 [loggers.py:320] Engine 000: Avg prompt throughput: 6.9 tokens/s, Avg generation throughput: 65.0 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 242 |
+
(APIServer pid=36395) INFO 11-19 03:38:34 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.54, Accepted throughput: 46.79 tokens/s, Drafted throughput: 55.19 tokens/s, Accepted: 468 tokens, Drafted: 552 tokens, Per-position acceptance rate: 0.951, 0.859, 0.734, Avg Draft acceptance rate: 84.8%
|
| 243 |
+
INFO: 127.0.0.1:57057 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 244 |
+
(APIServer pid=36395) INFO 11-19 03:38:44 [loggers.py:320] Engine 000: Avg prompt throughput: 5.0 tokens/s, Avg generation throughput: 61.6 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 245 |
+
(APIServer pid=36395) INFO 11-19 03:38:44 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.35, Accepted throughput: 43.29 tokens/s, Drafted throughput: 55.19 tokens/s, Accepted: 433 tokens, Drafted: 552 tokens, Per-position acceptance rate: 0.908, 0.766, 0.679, Avg Draft acceptance rate: 78.4%
|
| 246 |
+
INFO: 127.0.0.1:50893 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 247 |
+
(APIServer pid=36395) INFO 11-19 03:38:54 [loggers.py:320] Engine 000: Avg prompt throughput: 10.3 tokens/s, Avg generation throughput: 51.8 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 248 |
+
(APIServer pid=36395) INFO 11-19 03:38:54 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.83, Accepted throughput: 33.49 tokens/s, Drafted throughput: 54.89 tokens/s, Accepted: 335 tokens, Drafted: 549 tokens, Per-position acceptance rate: 0.809, 0.601, 0.421, Avg Draft acceptance rate: 61.0%
|
| 249 |
+
INFO: 127.0.0.1:39298 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 250 |
+
INFO: 127.0.0.1:22094 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 251 |
+
(APIServer pid=36395) INFO 11-19 03:39:04 [loggers.py:320] Engine 000: Avg prompt throughput: 9.0 tokens/s, Avg generation throughput: 56.9 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 252 |
+
(APIServer pid=36395) INFO 11-19 03:39:04 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.15, Accepted throughput: 38.89 tokens/s, Drafted throughput: 54.29 tokens/s, Accepted: 389 tokens, Drafted: 543 tokens, Per-position acceptance rate: 0.884, 0.702, 0.564, Avg Draft acceptance rate: 71.6%
|
| 253 |
+
INFO: 127.0.0.1:48039 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 254 |
+
(APIServer pid=36395) INFO 11-19 03:39:14 [loggers.py:320] Engine 000: Avg prompt throughput: 22.7 tokens/s, Avg generation throughput: 52.9 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 255 |
+
(APIServer pid=36395) INFO 11-19 03:39:14 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.89, Accepted throughput: 34.60 tokens/s, Drafted throughput: 54.90 tokens/s, Accepted: 346 tokens, Drafted: 549 tokens, Per-position acceptance rate: 0.863, 0.612, 0.415, Avg Draft acceptance rate: 63.0%
|
| 256 |
+
INFO: 127.0.0.1:36870 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 257 |
+
INFO: 127.0.0.1:16983 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 258 |
+
(APIServer pid=36395) INFO 11-19 03:39:24 [loggers.py:320] Engine 000: Avg prompt throughput: 9.4 tokens/s, Avg generation throughput: 55.0 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 259 |
+
(APIServer pid=36395) INFO 11-19 03:39:24 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.05, Accepted throughput: 37.10 tokens/s, Drafted throughput: 54.30 tokens/s, Accepted: 371 tokens, Drafted: 543 tokens, Per-position acceptance rate: 0.873, 0.674, 0.503, Avg Draft acceptance rate: 68.3%
|
| 260 |
+
INFO: 127.0.0.1:37639 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 261 |
+
INFO: 127.0.0.1:16872 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 262 |
+
(APIServer pid=36395) INFO 11-19 03:39:34 [loggers.py:320] Engine 000: Avg prompt throughput: 26.5 tokens/s, Avg generation throughput: 60.3 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 263 |
+
(APIServer pid=36395) INFO 11-19 03:39:34 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.35, Accepted throughput: 42.50 tokens/s, Drafted throughput: 54.29 tokens/s, Accepted: 425 tokens, Drafted: 543 tokens, Per-position acceptance rate: 0.901, 0.779, 0.669, Avg Draft acceptance rate: 78.3%
|
| 264 |
+
INFO: 127.0.0.1:24568 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 265 |
+
(APIServer pid=36395) INFO 11-19 03:39:44 [loggers.py:320] Engine 000: Avg prompt throughput: 3.4 tokens/s, Avg generation throughput: 61.9 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 266 |
+
(APIServer pid=36395) INFO 11-19 03:39:44 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.38, Accepted throughput: 43.49 tokens/s, Drafted throughput: 54.89 tokens/s, Accepted: 435 tokens, Drafted: 549 tokens, Per-position acceptance rate: 0.929, 0.776, 0.672, Avg Draft acceptance rate: 79.2%
|
| 267 |
+
INFO: 127.0.0.1:31936 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 268 |
+
(APIServer pid=36395) INFO 11-19 03:39:54 [loggers.py:320] Engine 000: Avg prompt throughput: 4.4 tokens/s, Avg generation throughput: 44.4 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 269 |
+
(APIServer pid=36395) INFO 11-19 03:39:54 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.42, Accepted throughput: 26.00 tokens/s, Drafted throughput: 54.90 tokens/s, Accepted: 260 tokens, Drafted: 549 tokens, Per-position acceptance rate: 0.732, 0.454, 0.235, Avg Draft acceptance rate: 47.4%
|
| 270 |
+
INFO: 127.0.0.1:32380 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 271 |
+
INFO: 127.0.0.1:20335 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 272 |
+
(APIServer pid=36395) INFO 11-19 03:40:04 [loggers.py:320] Engine 000: Avg prompt throughput: 13.5 tokens/s, Avg generation throughput: 53.2 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 273 |
+
(APIServer pid=36395) INFO 11-19 03:40:04 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.96, Accepted throughput: 35.40 tokens/s, Drafted throughput: 54.30 tokens/s, Accepted: 354 tokens, Drafted: 543 tokens, Per-position acceptance rate: 0.823, 0.635, 0.497, Avg Draft acceptance rate: 65.2%
|
| 274 |
+
INFO: 127.0.0.1:60648 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 275 |
+
(APIServer pid=36395) INFO 11-19 03:40:14 [loggers.py:320] Engine 000: Avg prompt throughput: 8.3 tokens/s, Avg generation throughput: 55.5 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 276 |
+
(APIServer pid=36395) INFO 11-19 03:40:14 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.01, Accepted throughput: 36.99 tokens/s, Drafted throughput: 55.19 tokens/s, Accepted: 370 tokens, Drafted: 552 tokens, Per-position acceptance rate: 0.826, 0.663, 0.522, Avg Draft acceptance rate: 67.0%
|
| 277 |
+
INFO: 127.0.0.1:23475 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 278 |
+
(APIServer pid=36395) INFO 11-19 03:40:24 [loggers.py:320] Engine 000: Avg prompt throughput: 3.4 tokens/s, Avg generation throughput: 48.3 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 279 |
+
(APIServer pid=36395) INFO 11-19 03:40:24 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.64, Accepted throughput: 30.09 tokens/s, Drafted throughput: 55.19 tokens/s, Accepted: 301 tokens, Drafted: 552 tokens, Per-position acceptance rate: 0.766, 0.527, 0.342, Avg Draft acceptance rate: 54.5%
|
| 280 |
+
INFO: 127.0.0.1:29718 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 281 |
+
(APIServer pid=36395) INFO 11-19 03:40:34 [loggers.py:320] Engine 000: Avg prompt throughput: 4.8 tokens/s, Avg generation throughput: 47.6 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 282 |
+
(APIServer pid=36395) INFO 11-19 03:40:34 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.58, Accepted throughput: 29.10 tokens/s, Drafted throughput: 55.19 tokens/s, Accepted: 291 tokens, Drafted: 552 tokens, Per-position acceptance rate: 0.777, 0.505, 0.299, Avg Draft acceptance rate: 52.7%
|
| 283 |
+
INFO: 127.0.0.1:34919 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 284 |
+
INFO: 127.0.0.1:51272 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 285 |
+
INFO: 127.0.0.1:39321 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 286 |
+
(APIServer pid=36395) INFO 11-19 03:40:44 [loggers.py:320] Engine 000: Avg prompt throughput: 92.1 tokens/s, Avg generation throughput: 56.3 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 287 |
+
(APIServer pid=36395) INFO 11-19 03:40:44 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.16, Accepted throughput: 38.50 tokens/s, Drafted throughput: 53.40 tokens/s, Accepted: 385 tokens, Drafted: 534 tokens, Per-position acceptance rate: 0.865, 0.725, 0.573, Avg Draft acceptance rate: 72.1%
|
| 288 |
+
INFO: 127.0.0.1:38335 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 289 |
+
INFO: 127.0.0.1:21210 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 290 |
+
INFO: 127.0.0.1:36945 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 291 |
+
(APIServer pid=36395) INFO 11-19 03:40:54 [loggers.py:320] Engine 000: Avg prompt throughput: 27.4 tokens/s, Avg generation throughput: 63.9 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 292 |
+
(APIServer pid=36395) INFO 11-19 03:40:54 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.59, Accepted throughput: 46.10 tokens/s, Drafted throughput: 53.40 tokens/s, Accepted: 461 tokens, Drafted: 534 tokens, Per-position acceptance rate: 0.955, 0.876, 0.758, Avg Draft acceptance rate: 86.3%
|
| 293 |
+
INFO: 127.0.0.1:40850 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 294 |
+
(APIServer pid=36395) INFO 11-19 03:41:04 [loggers.py:320] Engine 000: Avg prompt throughput: 3.3 tokens/s, Avg generation throughput: 55.6 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 295 |
+
(APIServer pid=36395) INFO 11-19 03:41:04 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.03, Accepted throughput: 37.40 tokens/s, Drafted throughput: 55.20 tokens/s, Accepted: 374 tokens, Drafted: 552 tokens, Per-position acceptance rate: 0.875, 0.674, 0.484, Avg Draft acceptance rate: 67.8%
|
| 296 |
+
INFO: 127.0.0.1:20891 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 297 |
+
(APIServer pid=36395) INFO 11-19 03:41:14 [loggers.py:320] Engine 000: Avg prompt throughput: 8.8 tokens/s, Avg generation throughput: 56.8 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 298 |
+
(APIServer pid=36395) INFO 11-19 03:41:14 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.09, Accepted throughput: 38.40 tokens/s, Drafted throughput: 55.20 tokens/s, Accepted: 384 tokens, Drafted: 552 tokens, Per-position acceptance rate: 0.853, 0.679, 0.554, Avg Draft acceptance rate: 69.6%
|
| 299 |
+
INFO: 127.0.0.1:31694 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 300 |
+
(APIServer pid=36395) INFO 11-19 03:41:24 [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%
|
| 301 |
+
(APIServer pid=36395) INFO 11-19 03:41:24 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.27, Accepted throughput: 41.49 tokens/s, Drafted throughput: 54.89 tokens/s, Accepted: 415 tokens, Drafted: 549 tokens, Per-position acceptance rate: 0.896, 0.749, 0.623, Avg Draft acceptance rate: 75.6%
|
| 302 |
+
INFO: 127.0.0.1:33266 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 303 |
+
(APIServer pid=36395) INFO 11-19 03:41:34 [loggers.py:320] Engine 000: Avg prompt throughput: 3.4 tokens/s, Avg generation throughput: 49.9 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 304 |
+
(APIServer pid=36395) INFO 11-19 03:41:34 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.73, Accepted throughput: 31.60 tokens/s, Drafted throughput: 54.90 tokens/s, Accepted: 316 tokens, Drafted: 549 tokens, Per-position acceptance rate: 0.754, 0.552, 0.421, Avg Draft acceptance rate: 57.6%
|
| 305 |
+
INFO: 127.0.0.1:19283 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 306 |
+
INFO: 127.0.0.1:62369 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 307 |
+
INFO: 127.0.0.1:42155 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 308 |
+
(APIServer pid=36395) INFO 11-19 03:41:44 [loggers.py:320] Engine 000: Avg prompt throughput: 48.8 tokens/s, Avg generation throughput: 60.0 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 309 |
+
(APIServer pid=36395) INFO 11-19 03:41:44 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.37, Accepted throughput: 42.20 tokens/s, Drafted throughput: 53.40 tokens/s, Accepted: 422 tokens, Drafted: 534 tokens, Per-position acceptance rate: 0.904, 0.792, 0.674, Avg Draft acceptance rate: 79.0%
|
| 310 |
+
INFO: 127.0.0.1:29584 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 311 |
+
(APIServer pid=36395) INFO 11-19 03:41:54 [loggers.py:320] Engine 000: Avg prompt throughput: 8.1 tokens/s, Avg generation throughput: 61.8 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 312 |
+
(APIServer pid=36395) INFO 11-19 03:41:54 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.34, Accepted throughput: 43.30 tokens/s, Drafted throughput: 55.49 tokens/s, Accepted: 433 tokens, Drafted: 555 tokens, Per-position acceptance rate: 0.914, 0.784, 0.643, Avg Draft acceptance rate: 78.0%
|
| 313 |
+
INFO: 127.0.0.1:47510 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 314 |
+
(APIServer pid=36395) INFO 11-19 03:42:04 [loggers.py:320] Engine 000: Avg prompt throughput: 2.4 tokens/s, Avg generation throughput: 60.0 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 315 |
+
(APIServer pid=36395) INFO 11-19 03:42:04 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.27, Accepted throughput: 41.60 tokens/s, Drafted throughput: 54.89 tokens/s, Accepted: 416 tokens, Drafted: 549 tokens, Per-position acceptance rate: 0.885, 0.754, 0.634, Avg Draft acceptance rate: 75.8%
|
| 316 |
+
INFO: 127.0.0.1:36948 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 317 |
+
(APIServer pid=36395) INFO 11-19 03:42:14 [loggers.py:320] Engine 000: Avg prompt throughput: 3.4 tokens/s, Avg generation throughput: 46.2 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 318 |
+
(APIServer pid=36395) INFO 11-19 03:42:14 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.52, Accepted throughput: 27.90 tokens/s, Drafted throughput: 54.89 tokens/s, Accepted: 279 tokens, Drafted: 549 tokens, Per-position acceptance rate: 0.716, 0.503, 0.306, Avg Draft acceptance rate: 50.8%
|
| 319 |
+
INFO: 127.0.0.1:64634 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 320 |
+
INFO: 127.0.0.1:55636 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 321 |
+
(APIServer pid=36395) INFO 11-19 03:42:24 [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%
|
| 322 |
+
(APIServer pid=36395) INFO 11-19 03:42:24 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.62, Accepted throughput: 29.30 tokens/s, Drafted throughput: 54.30 tokens/s, Accepted: 293 tokens, Drafted: 543 tokens, Per-position acceptance rate: 0.773, 0.519, 0.326, Avg Draft acceptance rate: 54.0%
|
| 323 |
+
INFO: 127.0.0.1:32255 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 324 |
+
INFO: 127.0.0.1:48185 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 325 |
+
(APIServer pid=36395) INFO 11-19 03:42:34 [loggers.py:320] Engine 000: Avg prompt throughput: 12.1 tokens/s, Avg generation throughput: 63.1 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 326 |
+
(APIServer pid=36395) INFO 11-19 03:42:34 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.47, Accepted throughput: 44.89 tokens/s, Drafted throughput: 54.59 tokens/s, Accepted: 449 tokens, Drafted: 546 tokens, Per-position acceptance rate: 0.918, 0.819, 0.731, Avg Draft acceptance rate: 82.2%
|
| 327 |
+
INFO: 127.0.0.1:34014 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 328 |
+
INFO: 127.0.0.1:65534 - "GET /metrics HTTP/1.1" 200 OK
|
| 329 |
+
INFO: 127.0.0.1:49917 - "GET /metrics HTTP/1.1" 200 OK
|
| 330 |
+
INFO: 127.0.0.1:18350 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 331 |
+
INFO: 127.0.0.1:24590 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 332 |
+
(APIServer pid=36395) INFO 11-19 03:42:44 [loggers.py:320] Engine 000: Avg prompt throughput: 37.6 tokens/s, Avg generation throughput: 64.1 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 333 |
+
(APIServer pid=36395) INFO 11-19 03:42:44 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.59, Accepted throughput: 46.30 tokens/s, Drafted throughput: 53.70 tokens/s, Accepted: 463 tokens, Drafted: 537 tokens, Per-position acceptance rate: 0.922, 0.866, 0.799, Avg Draft acceptance rate: 86.2%
|
| 334 |
+
INFO: 127.0.0.1:33720 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 335 |
+
INFO: 127.0.0.1:19310 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 336 |
+
(APIServer pid=36395) INFO 11-19 03:42:54 [loggers.py:320] Engine 000: Avg prompt throughput: 33.2 tokens/s, Avg generation throughput: 68.9 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 337 |
+
(APIServer pid=36395) INFO 11-19 03:42:54 [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: 0.983, 0.934, 0.884, Avg Draft acceptance rate: 93.4%
|
| 338 |
+
INFO: 127.0.0.1:59634 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 339 |
+
INFO: 127.0.0.1:47831 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 340 |
+
(APIServer pid=36395) INFO 11-19 03:43:04 [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%
|
| 341 |
+
(APIServer pid=36395) INFO 11-19 03:43:04 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.97, Accepted throughput: 53.40 tokens/s, Drafted throughput: 54.00 tokens/s, Accepted: 534 tokens, Drafted: 540 tokens, Per-position acceptance rate: 1.000, 0.983, 0.983, Avg Draft acceptance rate: 98.9%
|
| 342 |
+
INFO: 127.0.0.1:32306 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 343 |
+
INFO: 127.0.0.1:39490 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 344 |
+
(APIServer pid=36395) INFO 11-19 03:43:14 [loggers.py:320] Engine 000: Avg prompt throughput: 35.0 tokens/s, Avg generation throughput: 67.8 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 345 |
+
(APIServer pid=36395) INFO 11-19 03:43:14 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.76, Accepted throughput: 49.99 tokens/s, Drafted throughput: 54.29 tokens/s, Accepted: 500 tokens, Drafted: 543 tokens, Per-position acceptance rate: 0.978, 0.906, 0.878, Avg Draft acceptance rate: 92.1%
|
| 346 |
+
INFO: 127.0.0.1:25697 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 347 |
+
INFO: 127.0.0.1:56346 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 348 |
+
INFO: 127.0.0.1:23750 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 349 |
+
INFO: 127.0.0.1:18165 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 350 |
+
(APIServer pid=36395) INFO 11-19 03:43:24 [loggers.py:320] Engine 000: Avg prompt throughput: 69.2 tokens/s, Avg generation throughput: 66.7 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 351 |
+
(APIServer pid=36395) INFO 11-19 03:43:24 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.79, Accepted throughput: 49.10 tokens/s, Drafted throughput: 52.80 tokens/s, Accepted: 491 tokens, Drafted: 528 tokens, Per-position acceptance rate: 0.966, 0.926, 0.898, Avg Draft acceptance rate: 93.0%
|
| 352 |
+
INFO: 127.0.0.1:56363 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 353 |
+
(APIServer pid=36395) INFO 11-19 03:43:34 [loggers.py:320] Engine 000: Avg prompt throughput: 29.9 tokens/s, Avg generation throughput: 71.2 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 354 |
+
(APIServer pid=36395) INFO 11-19 03:43:34 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.91, Accepted throughput: 52.89 tokens/s, Drafted throughput: 54.59 tokens/s, Accepted: 529 tokens, Drafted: 546 tokens, Per-position acceptance rate: 0.995, 0.973, 0.940, Avg Draft acceptance rate: 96.9%
|
| 355 |
+
INFO: 127.0.0.1:40942 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 356 |
+
INFO: 127.0.0.1:22848 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 357 |
+
INFO: 127.0.0.1:30471 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 358 |
+
(APIServer pid=36395) INFO 11-19 03:43:44 [loggers.py:320] Engine 000: Avg prompt throughput: 71.0 tokens/s, Avg generation throughput: 65.0 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 359 |
+
(APIServer pid=36395) INFO 11-19 03:43:44 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.66, Accepted throughput: 47.39 tokens/s, Drafted throughput: 53.39 tokens/s, Accepted: 474 tokens, Drafted: 534 tokens, Per-position acceptance rate: 0.944, 0.904, 0.815, Avg Draft acceptance rate: 88.8%
|
| 360 |
+
INFO: 127.0.0.1:60264 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 361 |
+
INFO: 127.0.0.1:51056 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 362 |
+
(APIServer pid=36395) INFO 11-19 03:43:54 [loggers.py:320] Engine 000: Avg prompt throughput: 17.5 tokens/s, Avg generation throughput: 69.6 tokens/s, Running: 0 reqs, Waiting: 0 reqs, GPU KV cache usage: 0.0%
|
| 363 |
+
(APIServer pid=36395) INFO 11-19 03:43:54 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.81, Accepted throughput: 51.40 tokens/s, Drafted throughput: 54.90 tokens/s, Accepted: 514 tokens, Drafted: 549 tokens, Per-position acceptance rate: 0.962, 0.951, 0.896, Avg Draft acceptance rate: 93.6%
|
| 364 |
+
INFO: 127.0.0.1:28691 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 365 |
+
INFO: 127.0.0.1:35161 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 366 |
+
(APIServer pid=36395) INFO 11-19 03:44:04 [loggers.py:320] Engine 000: Avg prompt throughput: 47.8 tokens/s, Avg generation throughput: 65.7 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 367 |
+
(APIServer pid=36395) INFO 11-19 03:44:04 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.66, Accepted throughput: 47.70 tokens/s, Drafted throughput: 53.69 tokens/s, Accepted: 477 tokens, Drafted: 537 tokens, Per-position acceptance rate: 0.978, 0.888, 0.799, Avg Draft acceptance rate: 88.8%
|
| 368 |
+
INFO: 127.0.0.1:23973 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 369 |
+
INFO: 127.0.0.1:42908 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 370 |
+
INFO: 127.0.0.1:20534 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 371 |
+
(APIServer pid=36395) INFO 11-19 03:44:14 [loggers.py:320] Engine 000: Avg prompt throughput: 53.8 tokens/s, Avg generation throughput: 70.0 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 372 |
+
(APIServer pid=36395) INFO 11-19 03:44:14 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.95, Accepted throughput: 52.49 tokens/s, Drafted throughput: 53.39 tokens/s, Accepted: 525 tokens, Drafted: 534 tokens, Per-position acceptance rate: 0.994, 0.983, 0.972, Avg Draft acceptance rate: 98.3%
|
| 373 |
+
INFO: 127.0.0.1:57717 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 374 |
+
INFO: 127.0.0.1:22392 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 375 |
+
(APIServer pid=36395) INFO 11-19 03:44:24 [loggers.py:320] Engine 000: Avg prompt throughput: 16.7 tokens/s, Avg generation throughput: 59.0 tokens/s, Running: 0 reqs, Waiting: 0 reqs, GPU KV cache usage: 0.0%
|
| 376 |
+
(APIServer pid=36395) INFO 11-19 03:44:24 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.23, Accepted throughput: 40.90 tokens/s, Drafted throughput: 54.90 tokens/s, Accepted: 409 tokens, Drafted: 549 tokens, Per-position acceptance rate: 0.891, 0.732, 0.612, Avg Draft acceptance rate: 74.5%
|
| 377 |
+
INFO: 127.0.0.1:23907 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 378 |
+
INFO: 127.0.0.1:20793 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 379 |
+
(APIServer pid=36395) INFO 11-19 03:44:34 [loggers.py:320] Engine 000: Avg prompt throughput: 56.0 tokens/s, Avg generation throughput: 67.0 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 380 |
+
(APIServer pid=36395) INFO 11-19 03:44:34 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.77, Accepted throughput: 49.29 tokens/s, Drafted throughput: 53.39 tokens/s, Accepted: 493 tokens, Drafted: 534 tokens, Per-position acceptance rate: 0.955, 0.916, 0.899, Avg Draft acceptance rate: 92.3%
|
| 381 |
+
INFO: 127.0.0.1:36875 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 382 |
+
INFO: 127.0.0.1:47719 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 383 |
+
INFO: 127.0.0.1:33988 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 384 |
+
INFO: 127.0.0.1:64196 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 385 |
+
(APIServer pid=36395) INFO 11-19 03:44:44 [loggers.py:320] Engine 000: Avg prompt throughput: 79.2 tokens/s, Avg generation throughput: 68.6 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 386 |
+
(APIServer pid=36395) INFO 11-19 03:44:44 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.91, Accepted throughput: 51.29 tokens/s, Drafted throughput: 52.79 tokens/s, Accepted: 513 tokens, Drafted: 528 tokens, Per-position acceptance rate: 0.994, 0.977, 0.943, Avg Draft acceptance rate: 97.2%
|
| 387 |
+
INFO: 127.0.0.1:57959 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 388 |
+
INFO: 127.0.0.1:33454 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 389 |
+
(APIServer pid=36395) INFO 11-19 03:44:54 [loggers.py:320] Engine 000: Avg prompt throughput: 40.9 tokens/s, Avg generation throughput: 66.2 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 390 |
+
(APIServer pid=36395) INFO 11-19 03:44:54 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.68, Accepted throughput: 48.50 tokens/s, Drafted throughput: 54.30 tokens/s, Accepted: 485 tokens, Drafted: 543 tokens, Per-position acceptance rate: 0.934, 0.895, 0.851, Avg Draft acceptance rate: 89.3%
|
| 391 |
+
INFO: 127.0.0.1:44806 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 392 |
+
INFO: 127.0.0.1:31033 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 393 |
+
INFO: 127.0.0.1:31109 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 394 |
+
(APIServer pid=36395) INFO 11-19 03:45:04 [loggers.py:320] Engine 000: Avg prompt throughput: 57.4 tokens/s, Avg generation throughput: 69.6 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 395 |
+
(APIServer pid=36395) INFO 11-19 03:45:04 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.94, Accepted throughput: 52.30 tokens/s, Drafted throughput: 53.40 tokens/s, Accepted: 523 tokens, Drafted: 534 tokens, Per-position acceptance rate: 1.000, 0.983, 0.955, Avg Draft acceptance rate: 97.9%
|
| 396 |
+
INFO: 127.0.0.1:61359 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 397 |
+
INFO: 127.0.0.1:39362 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 398 |
+
(APIServer pid=36395) INFO 11-19 03:45:14 [loggers.py:320] Engine 000: Avg prompt throughput: 37.7 tokens/s, Avg generation throughput: 65.2 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 399 |
+
(APIServer pid=36395) INFO 11-19 03:45:14 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.58, Accepted throughput: 46.99 tokens/s, Drafted throughput: 54.59 tokens/s, Accepted: 470 tokens, Drafted: 546 tokens, Per-position acceptance rate: 0.934, 0.863, 0.786, Avg Draft acceptance rate: 86.1%
|
| 400 |
+
INFO: 127.0.0.1:19346 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 401 |
+
INFO: 127.0.0.1:34832 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 402 |
+
INFO: 127.0.0.1:60156 - "GET /metrics HTTP/1.1" 200 OK
|
| 403 |
+
INFO: 127.0.0.1:27163 - "GET /metrics HTTP/1.1" 200 OK
|
| 404 |
+
INFO: 127.0.0.1:53920 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 405 |
+
(APIServer pid=36395) INFO 11-19 03:45:24 [loggers.py:320] Engine 000: Avg prompt throughput: 35.2 tokens/s, Avg generation throughput: 63.8 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 406 |
+
(APIServer pid=36395) INFO 11-19 03:45:24 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.58, Accepted throughput: 45.99 tokens/s, Drafted throughput: 53.39 tokens/s, Accepted: 460 tokens, Drafted: 534 tokens, Per-position acceptance rate: 0.927, 0.848, 0.809, Avg Draft acceptance rate: 86.1%
|
| 407 |
+
INFO: 127.0.0.1:27494 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 408 |
+
INFO: 127.0.0.1:43998 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 409 |
+
(APIServer pid=36395) INFO 11-19 03:45:34 [loggers.py:320] Engine 000: Avg prompt throughput: 12.1 tokens/s, Avg generation throughput: 68.9 tokens/s, Running: 0 reqs, Waiting: 0 reqs, GPU KV cache usage: 0.0%
|
| 410 |
+
(APIServer pid=36395) INFO 11-19 03:45:34 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.77, Accepted throughput: 50.89 tokens/s, Drafted throughput: 55.19 tokens/s, Accepted: 509 tokens, Drafted: 552 tokens, Per-position acceptance rate: 0.978, 0.935, 0.853, Avg Draft acceptance rate: 92.2%
|
| 411 |
+
INFO: 127.0.0.1:48540 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 412 |
+
INFO: 127.0.0.1:46775 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 413 |
+
(APIServer pid=36395) INFO 11-19 03:45:44 [loggers.py:320] Engine 000: Avg prompt throughput: 28.9 tokens/s, Avg generation throughput: 63.8 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 414 |
+
(APIServer pid=36395) INFO 11-19 03:45:44 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.58, Accepted throughput: 46.09 tokens/s, Drafted throughput: 53.69 tokens/s, Accepted: 461 tokens, Drafted: 537 tokens, Per-position acceptance rate: 0.955, 0.849, 0.771, Avg Draft acceptance rate: 85.8%
|
| 415 |
+
INFO: 127.0.0.1:52963 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 416 |
+
INFO: 127.0.0.1:57633 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 417 |
+
(APIServer pid=36395) INFO 11-19 03:45:54 [loggers.py:320] Engine 000: Avg prompt throughput: 21.6 tokens/s, Avg generation throughput: 64.6 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 418 |
+
(APIServer pid=36395) INFO 11-19 03:45:54 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.59, Accepted throughput: 46.69 tokens/s, Drafted throughput: 53.99 tokens/s, Accepted: 467 tokens, Drafted: 540 tokens, Per-position acceptance rate: 0.939, 0.867, 0.789, Avg Draft acceptance rate: 86.5%
|
| 419 |
+
INFO: 127.0.0.1:44861 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 420 |
+
INFO: 127.0.0.1:53231 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 421 |
+
(APIServer pid=36395) INFO 11-19 03:46:04 [loggers.py:320] Engine 000: Avg prompt throughput: 23.3 tokens/s, Avg generation throughput: 68.4 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 422 |
+
(APIServer pid=36395) INFO 11-19 03:46:04 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.77, Accepted throughput: 50.20 tokens/s, Drafted throughput: 54.30 tokens/s, Accepted: 502 tokens, Drafted: 543 tokens, Per-position acceptance rate: 0.983, 0.939, 0.851, Avg Draft acceptance rate: 92.4%
|
| 423 |
+
INFO: 127.0.0.1:27624 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 424 |
+
INFO: 127.0.0.1:18212 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 425 |
+
(APIServer pid=36395) INFO 11-19 03:46:14 [loggers.py:320] Engine 000: Avg prompt throughput: 22.0 tokens/s, Avg generation throughput: 66.2 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 426 |
+
(APIServer pid=36395) INFO 11-19 03:46:14 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.65, Accepted throughput: 48.30 tokens/s, Drafted throughput: 54.59 tokens/s, Accepted: 483 tokens, Drafted: 546 tokens, Per-position acceptance rate: 0.973, 0.874, 0.808, Avg Draft acceptance rate: 88.5%
|
| 427 |
+
INFO: 127.0.0.1:51010 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 428 |
+
INFO: 127.0.0.1:31460 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 429 |
+
(APIServer pid=36395) INFO 11-19 03:46:24 [loggers.py:320] Engine 000: Avg prompt throughput: 19.7 tokens/s, Avg generation throughput: 64.2 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 430 |
+
(APIServer pid=36395) INFO 11-19 03:46:24 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.56, Accepted throughput: 46.40 tokens/s, Drafted throughput: 54.30 tokens/s, Accepted: 464 tokens, Drafted: 543 tokens, Per-position acceptance rate: 0.945, 0.856, 0.762, Avg Draft acceptance rate: 85.5%
|
| 431 |
+
INFO: 127.0.0.1:45839 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 432 |
+
INFO: 127.0.0.1:43688 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 433 |
+
(APIServer pid=36395) INFO 11-19 03:46:34 [loggers.py:320] Engine 000: Avg prompt throughput: 24.0 tokens/s, Avg generation throughput: 68.4 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 434 |
+
(APIServer pid=36395) INFO 11-19 03:46:34 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.78, Accepted throughput: 50.39 tokens/s, Drafted throughput: 54.29 tokens/s, Accepted: 504 tokens, Drafted: 543 tokens, Per-position acceptance rate: 0.972, 0.939, 0.873, Avg Draft acceptance rate: 92.8%
|
runs/r3/logs/eval_main.log
CHANGED
|
@@ -6,3 +6,6 @@
|
|
| 6 |
PASS
|
| 7 |
[03:33:39] repeat 1/1: fresh vLLM process
|
| 8 |
[03:33:39] starting vLLM (log: /workspace/emc/work/logs/20261119-033339-eval-r1.log)
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
PASS
|
| 7 |
[03:33:39] repeat 1/1: fresh vLLM process
|
| 8 |
[03:33:39] starting vLLM (log: /workspace/emc/work/logs/20261119-033339-eval-r1.log)
|
| 9 |
+
[03:37:05] vLLM ready after 206s
|
| 10 |
+
[03:42:41] chat: 40 prompts done
|
| 11 |
+
[03:45:24] code: 40 prompts done
|
runs/r3/logs/stats.log
CHANGED
|
@@ -42,3 +42,4 @@
|
|
| 42 |
[2026-11-19 03:33:26] 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%, 64 W | load 0.0 | largest python 0.1 GB, python 0.1 GB, python 0.0 GB
|
| 43 |
[2026-11-19 03:33:29] 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%, 63 W | load 0.0 | largest python 0.1 GB, python 0.1 GB, python 0.0 GB
|
| 44 |
[2026-11-19 03:36:32] ram 3.8/172 GB used (cache 0.3, shmem 0.0) | cgroup 3.8/172 GB | disk 143 GB used | gpu 53.7/95.6 GiB, 0%, 79 W | load 0.0 | largest VLLM::EngineCor 3.3 GB, vllm 2.4 GB, python 1.2 GB
|
|
|
|
|
|
| 42 |
[2026-11-19 03:33:26] 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%, 64 W | load 0.0 | largest python 0.1 GB, python 0.1 GB, python 0.0 GB
|
| 43 |
[2026-11-19 03:33:29] 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%, 63 W | load 0.0 | largest python 0.1 GB, python 0.1 GB, python 0.0 GB
|
| 44 |
[2026-11-19 03:36:32] ram 3.8/172 GB used (cache 0.3, shmem 0.0) | cgroup 3.8/172 GB | disk 143 GB used | gpu 53.7/95.6 GiB, 0%, 79 W | load 0.0 | largest VLLM::EngineCor 3.3 GB, vllm 2.4 GB, python 1.2 GB
|
| 45 |
+
[2026-11-19 03:46:34] ram 4.6/172 GB used (cache 0.3, shmem 0.0) | cgroup 4.6/172 GB | disk 143 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
|
@@ -36,3 +36,4 @@ No files have been modified since last commit. Skipping to prevent empty commit.
|
|
| 36 |
uploaded sub_bf16 to mudassiruddin21/EMC:submissions/r3/main3-bf16 (https://huggingface.co/datasets/mudassiruddin21/EMC/commit/51e6f3c7557c7023ffa591c3fd11169548573f13)
|
| 37 |
No files have been modified since last commit. Skipping to prevent empty commit.
|
| 38 |
uploaded sub_fp8 to mudassiruddin21/EMC:submissions/r3/main3-fp8 (https://huggingface.co/datasets/mudassiruddin21/EMC/commit/51e6f3c7557c7023ffa591c3fd11169548573f13)
|
|
|
|
|
|
| 36 |
uploaded sub_bf16 to mudassiruddin21/EMC:submissions/r3/main3-bf16 (https://huggingface.co/datasets/mudassiruddin21/EMC/commit/51e6f3c7557c7023ffa591c3fd11169548573f13)
|
| 37 |
No files have been modified since last commit. Skipping to prevent empty commit.
|
| 38 |
uploaded sub_fp8 to mudassiruddin21/EMC:submissions/r3/main3-fp8 (https://huggingface.co/datasets/mudassiruddin21/EMC/commit/51e6f3c7557c7023ffa591c3fd11169548573f13)
|
| 39 |
+
uploaded logs/results to mudassiruddin21/EMC:runs/r3
|