Run r3: logs and results
Browse files- runs/r3/engine_logs/20261119-000859-eval-r1.log +246 -0
- runs/r3/logs/pipeline.log +1 -0
- runs/r3/logs/pipeline.out +1 -0
- runs/r3/logs/smoke_fp8.log +10 -0
- runs/r3/logs/stats.log +2 -0
- runs/r3/logs/sync.log +1 -0
runs/r3/engine_logs/20261119-000859-eval-r1.log
ADDED
|
@@ -0,0 +1,246 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
vllm serve /workspace/emc/models/target --served-model-name target --host 127.0.0.1 --port 53752 --dtype bfloat16 --max-model-len 16384 --gpu-memory-utilization 0.9 --max-num-seqs 4 --seed 0 --no-enable-prefix-caching --generation-config vllm --language-model-only --speculative-config {"method": "mtp", "model": "/workspace/emc/models/probe-fp8", "num_speculative_tokens": 3}
|
| 2 |
+
|
| 3 |
+
(APIServer pid=14373) INFO 11-19 00:09:18 [api_utils.py:347]
|
| 4 |
+
(APIServer pid=14373) INFO 11-19 00:09:18 [api_utils.py:347] █ █ █▄ ▄█
|
| 5 |
+
(APIServer pid=14373) INFO 11-19 00:09:18 [api_utils.py:347] ▄▄ ▄█ █ █ █ ▀▄▀ █ version 0.31.0
|
| 6 |
+
(APIServer pid=14373) INFO 11-19 00:09:18 [api_utils.py:347] █▄█▀ █ █ █ █ model /workspace/emc/models/target
|
| 7 |
+
(APIServer pid=14373) INFO 11-19 00:09:18 [api_utils.py:347] ▀▀ ▀▀▀▀▀ ▀▀▀▀▀ ▀ ▀
|
| 8 |
+
(APIServer pid=14373) INFO 11-19 00:09:18 [api_utils.py:347]
|
| 9 |
+
(APIServer pid=14373) INFO 11-19 00:09:18 [api_utils.py:286] non-default args: {'model_tag': '/workspace/emc/models/target', 'host': '127.0.0.1', 'port': 53752, 'model': '/workspace/emc/models/target', 'dtype': 'bfloat16', 'max_model_len': 16384, 'served_model_name': ['target'], 'generation_config': 'vllm', 'gpu_memory_utilization': 0.9, 'enable_prefix_caching': False, 'language_model_only': True, 'max_num_seqs': 4, 'speculative_config': {'method': 'mtp', 'model': '/workspace/emc/models/probe-fp8', 'num_speculative_tokens': 3}}
|
| 10 |
+
(APIServer pid=14373) INFO 11-19 00:09:36 [model.py:699] Resolved architecture: Qwen3_5ForConditionalGeneration
|
| 11 |
+
(APIServer pid=14373) INFO 11-19 00:09:36 [model.py:2109] Using max model len 16384
|
| 12 |
+
(APIServer pid=14373) INFO 11-19 00:09:38 [model.py:917] All limits of multimodal modalities supported by the model are set to 0, running in text-only mode.
|
| 13 |
+
(APIServer pid=14373) INFO 11-19 00:09:38 [model.py:974] Disabled mm_prefix attention mode because multimodal inputs are configuration-disabled. Attention backends without mm_prefix support may now be selected.
|
| 14 |
+
(APIServer pid=14373) INFO 11-19 00:09:57 [model.py:699] Resolved architecture: Qwen3_5MTP
|
| 15 |
+
(APIServer pid=14373) INFO 11-19 00:09:57 [model.py:2109] Using max model len 262144
|
| 16 |
+
(APIServer pid=14373) WARNING 11-19 00:09:57 [speculative.py:1375] Enabling num_speculative_tokens > 1 will run multiple times of forward on same MTP layer,which may result in lower acceptance rate
|
| 17 |
+
(APIServer pid=14373) INFO 11-19 00:09:57 [speculative.py:1663] Overriding draft model max model len from 262144 to 16384
|
| 18 |
+
(APIServer pid=14373) INFO 11-19 00:09:57 [scheduler.py:308] Chunked prefill is enabled with max_num_batched_tokens=8192.
|
| 19 |
+
(APIServer pid=14373) INFO 11-19 00:09:57 [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'])
|
| 20 |
+
(EngineCore pid=14912) INFO 11-19 00:10:15 [core.py:129] Initializing a V1 LLM engine (v0.31.0) with config: model='/workspace/emc/models/target', speculative_config=SpeculativeConfig(method='mtp', model='/workspace/emc/models/probe-fp8', num_spec_tokens=3), tokenizer='/workspace/emc/models/target', skip_tokenizer_init=False, tokenizer_mode=auto, revision=None, tokenizer_revision=None, trust_remote_code=False, dtype=torch.bfloat16, max_seq_len=16384, download_dir=None, load_format=auto, tensor_parallel_size=1, pipeline_parallel_size=1, data_parallel_size=1, decode_context_parallel_size=1, dcp_comm_backend=ag_rs, disable_custom_all_reduce=False, quantization=None, quantization_config=None, enforce_eager=False, aux_output_config=AuxOutputConfig(enable_return_routed_experts=False, max_bytes=None), kv_cache_dtype=auto, device_config=cuda, structured_outputs_config=StructuredOutputsConfig(backend='auto', disable_any_whitespace=False, disable_additional_properties=False, reasoning_parser='', reasoning_parser_plugin='', enable_in_reasoning=False), observability_config=ObservabilityConfig(show_hidden_metrics_for_version=None, otlp_traces_endpoint=None, collect_detailed_traces=None, per_request_spec_decode_metrics='none', kv_cache_metrics=False, kv_cache_metrics_sample=0.01, cudagraph_metrics=False, enable_layerwise_nvtx_tracing=False, enable_mfu_metrics=False, enable_mm_processor_stats=False, enable_logging_iteration_details=False, jit_monitor_mode='warn', jit_monitor_verbose=False), seed=0, served_model_name=target, enable_prefix_caching=False, enable_chunked_prefill=True, pooler_config=None, compilation_config={'mode': <CompilationMode.VLLM_COMPILE: 3>, 'debug_dump_path': None, 'cache_dir': '', 'compile_cache_save_format': 'binary', 'backend': 'inductor', 'custom_ops': ['none'], 'ir_enable_torch_wrap': True, 'splitting_ops': ['vllm::unified_attention_with_output', 'vllm::unified_mla_attention_with_output', 'vllm::mamba_mixer2', 'vllm::mamba_mixer', 'vllm::short_conv', 'vllm::qwen4_exp_ple_short_conv', 'vllm::qwen4_exp_qsa_with_output', 'vllm::linear_attention', 'vllm::qwen_gdn_attention_core', 'vllm::qwen_gdn_attention_core_fused_norm_packed', 'vllm::gdn_attention_core_xpu', 'vllm::olmo_hybrid_gdn_full_forward', 'vllm::sparse_attn_indexer', 'vllm::rocm_aiter_sparse_attn_indexer', 'vllm::deepseek_v4_attention', 'vllm::hpc_rope_norm_forward', 'vllm::unified_kv_cache_update', 'vllm::unified_mla_kv_cache_update'], 'compile_mm_encoder': False, 'cudagraph_mm_encoder': False, 'encoder_cudagraph_token_budgets': [], 'encoder_cudagraph_max_vision_items_per_batch': 0, 'encoder_cudagraph_max_frames_per_batch': None, 'compile_sizes': [], 'compile_ranges_endpoints': [8192], 'inductor_compile_config': {'enable_auto_functionalized_v2': False, 'combo_kernels': True, 'benchmark_combo_kernel': True}, 'inductor_passes': {}, 'cudagraph_mode': <CUDAGraphMode.FULL_AND_PIECEWISE: (2, 1)>, 'cudagraph_num_of_warmups': 1, 'cudagraph_capture_sizes': [1, 2, 4, 8, 16, 24, 32], 'cudagraph_copy_inputs': False, 'cudagraph_specialize_lora': True, 'use_inductor_graph_partition': False, 'pass_config': {'fuse_norm_quant': False, 'fuse_act_quant': False, 'fuse_attn_quant': False, 'enable_sp': False, 'fuse_gemm_comms': False, 'fuse_allreduce_rms': False, 'enable_qk_norm_rope_fusion': False, 'fuse_rope_kvcache_cat_mla': False, 'fuse_act_padding': False, 'fuse_qk_norm_rope_kvcache': False}, 'max_cudagraph_capture_size': 32, 'dynamic_shapes_config': {'type': <DynamicShapesType.BACKED: 'backed'>, 'evaluate_guards': False, 'assume_32_bit_indexing': False}, 'local_cache_dir': None, 'fast_moe_cold_start': False, 'static_all_moe_layers': []}, kernel_config=KernelConfig(ir_op_priority=IrOpPriorityConfig(rms_norm=['native'], fused_add_rms_norm=['native'], gelu_and_mul_sparse=['triton', 'native']), enable_flashinfer_autotune=True, enable_cutedsl_warmup=True, enable_jit_warmup=True, moe_backend='auto', sparse_indexer_topk_backend='auto', linear_backend='auto', linear_backend_per_quant=None)
|
| 21 |
+
(EngineCore pid=14912) INFO 11-19 00:10:17 [parallel_state.py:1825] world_size=1 rank=0 local_rank=0 distributed_init_method=file:///workspace/emc/tmp/vllm_dist_dbdbc9c770f34419a2fb7d0c49b0cb31 backend=nccl
|
| 22 |
+
(EngineCore pid=14912) INFO 11-19 00:10:17 [parallel_state.py:2269] rank 0 in world size 1 is assigned as DP rank 0, PP rank 0, PCP rank 0, TP rank 0, ETP rank 0, EP rank N/A, EPLB rank N/A
|
| 23 |
+
(EngineCore pid=14912) INFO 11-19 00:10:17 [gpu_worker.py:479] Using V2 Model Runner
|
| 24 |
+
(EngineCore pid=14912) INFO 11-19 00:10:18 [model_runner.py:375] Loading model from scratch...
|
| 25 |
+
(EngineCore pid=14912) INFO 11-19 00:10:19 [cuda.py:585] Using backend AttentionBackendEnum.FLASH_ATTN for vit attention
|
| 26 |
+
(EngineCore pid=14912) INFO 11-19 00:10:19 [mm_encoder_attention.py:372] Using AttentionBackendEnum.FLASH_ATTN for MMEncoderAttention.
|
| 27 |
+
(EngineCore pid=14912) INFO 11-19 00:10:19 [qwen_gdn_linear_attn.py:177] Using FlashInfer GDN prefill kernel (requested=auto, head_k_dim=128).
|
| 28 |
+
(EngineCore pid=14912) INFO 11-19 00:10:19 [qwen_gdn_linear_attn.py:533] GDN decode kernel: cuda
|
| 29 |
+
(EngineCore pid=14912) INFO 11-19 00:10:21 [cuda.py:526] Using FLASH_ATTN attention backend out of potential backends: ['FLASH_ATTN', 'FLASHINFER', 'TRITON_ATTN', 'FLEX_ATTENTION'].
|
| 30 |
+
(EngineCore pid=14912) INFO 11-19 00:10:21 [flash_attn.py:1167] Using FlashAttention version 2
|
| 31 |
+
(EngineCore pid=14912) INFO 11-19 00:10:22 [weight_utils.py:1063] Filesystem type for checkpoints: OVERLAY. Checkpoint size: 51.75 GiB. Available RAM: 156.18 GiB.
|
| 32 |
+
(EngineCore pid=14912) INFO 11-19 00:10:22 [weight_utils.py:1086] Auto-prefetch is disabled because the filesystem (OVERLAY) is not a recognized network FS (NFS/Lustre). If you want to force prefetching, start vLLM with --safetensors-load-strategy=prefetch.
|
| 33 |
+
|
| 34 |
+
Loading safetensors checkpoint shards: 0% Completed | 0/18 [00:00<?, ?it/s]
|
| 35 |
+
|
| 36 |
+
Loading safetensors checkpoint shards: 6% Completed | 1/18 [00:00<00:07, 2.28it/s]
|
| 37 |
+
|
| 38 |
+
Loading safetensors checkpoint shards: 11% Completed | 2/18 [00:00<00:06, 2.36it/s]
|
| 39 |
+
|
| 40 |
+
Loading safetensors checkpoint shards: 17% Completed | 3/18 [00:01<00:05, 2.61it/s]
|
| 41 |
+
|
| 42 |
+
Loading safetensors checkpoint shards: 22% Completed | 4/18 [00:01<00:06, 2.28it/s]
|
| 43 |
+
|
| 44 |
+
Loading safetensors checkpoint shards: 28% Completed | 5/18 [00:01<00:04, 2.66it/s]
|
| 45 |
+
|
| 46 |
+
Loading safetensors checkpoint shards: 33% Completed | 6/18 [00:02<00:05, 2.34it/s]
|
| 47 |
+
|
| 48 |
+
Loading safetensors checkpoint shards: 39% Completed | 7/18 [00:02<00:04, 2.71it/s]
|
| 49 |
+
|
| 50 |
+
Loading safetensors checkpoint shards: 44% Completed | 8/18 [00:03<00:04, 2.47it/s]
|
| 51 |
+
|
| 52 |
+
Loading safetensors checkpoint shards: 50% Completed | 9/18 [00:03<00:03, 2.78it/s]
|
| 53 |
+
|
| 54 |
+
Loading safetensors checkpoint shards: 56% Completed | 10/18 [00:03<00:03, 2.50it/s]
|
| 55 |
+
|
| 56 |
+
Loading safetensors checkpoint shards: 61% Completed | 11/18 [00:04<00:02, 2.82it/s]
|
| 57 |
+
|
| 58 |
+
Loading safetensors checkpoint shards: 67% Completed | 12/18 [00:04<00:02, 2.50it/s]
|
| 59 |
+
|
| 60 |
+
Loading safetensors checkpoint shards: 72% Completed | 13/18 [00:04<00:01, 2.79it/s]
|
| 61 |
+
|
| 62 |
+
Loading safetensors checkpoint shards: 78% Completed | 14/18 [00:05<00:01, 2.45it/s]
|
| 63 |
+
|
| 64 |
+
Loading safetensors checkpoint shards: 83% Completed | 15/18 [00:05<00:01, 2.74it/s]
|
| 65 |
+
|
| 66 |
+
Loading safetensors checkpoint shards: 89% Completed | 16/18 [00:06<00:00, 2.52it/s]
|
| 67 |
+
|
| 68 |
+
Loading safetensors checkpoint shards: 94% Completed | 17/18 [00:06<00:00, 2.81it/s]
|
| 69 |
+
|
| 70 |
+
Loading safetensors checkpoint shards: 100% Completed | 18/18 [00:06<00:00, 2.91it/s]
|
| 71 |
+
|
| 72 |
+
Loading safetensors checkpoint shards: 100% Completed | 18/18 [00:06<00:00, 2.63it/s]
|
| 73 |
+
|
| 74 |
+
(EngineCore pid=14912) INFO 11-19 00:10:29 [default_loader.py:484] Loading weights took 6.86 seconds
|
| 75 |
+
(EngineCore pid=14912) INFO 11-19 00:10:29 [weight_utils.py:1063] Filesystem type for checkpoints: OVERLAY. Checkpoint size: 0.50 GiB. Available RAM: 156.90 GiB.
|
| 76 |
+
|
| 77 |
+
Loading safetensors checkpoint shards: 0% Completed | 0/1 [00:00<?, ?it/s]
|
| 78 |
+
|
| 79 |
+
Loading safetensors checkpoint shards: 100% Completed | 1/1 [00:00<00:00, 4.34it/s]
|
| 80 |
+
|
| 81 |
+
Loading safetensors checkpoint shards: 100% Completed | 1/1 [00:00<00:00, 4.34it/s]
|
| 82 |
+
|
| 83 |
+
(EngineCore pid=14912) INFO 11-19 00:10:29 [default_loader.py:484] Loading weights took 0.23 seconds
|
| 84 |
+
(EngineCore pid=14912) INFO 11-19 00:10:29 [model_runner.py:407] Model loading took 51.02 GiB memory and 11.569117 seconds
|
| 85 |
+
(EngineCore pid=14912) INFO 11-19 00:10:29 [topk_topp_sampler.py:96] FlashInfer top-p/top-k sampling disabled via VLLM_USE_FLASHINFER_SAMPLER=0.
|
| 86 |
+
(EngineCore pid=14912) WARNING 11-19 00:10:29 [torch_utils.py:274] OMP_NUM_THREADS=20 is set; leaving Torch threads at 20 for serving. Multi-threaded torch CPU ops during serving can degrade performance through spin-wait contention and cgroup CPU-quota throttling.
|
| 87 |
+
(EngineCore pid=14912) INFO 11-19 00:10:29 [interface.py:989] Setting attention block size to 800 tokens to ensure that attention page size is >= mamba page size.
|
| 88 |
+
(EngineCore pid=14912) INFO 11-19 00:10:29 [interface.py:1013] Padding mamba page size by 0.25% to ensure that mamba page size and attention page size are exactly equal.
|
| 89 |
+
(EngineCore pid=14912) INFO 11-19 00:10:29 [utils.py:343] Using LBNHC KV cache layout.
|
| 90 |
+
(EngineCore pid=14912) INFO 11-19 00:10:37 [backends.py:1090] Using cache directory: /workspace/emc/work/cache/vllm/torch_compile_cache/070dc5d8c5/rank_0_0/backbone for vLLM's torch.compile
|
| 91 |
+
(EngineCore pid=14912) INFO 11-19 00:10:37 [backends.py:1151] Dynamo bytecode transform time: 6.58 s
|
| 92 |
+
(EngineCore pid=14912) INFO 11-19 00:10:54 [backends.py:391] Compiling a graph for compile range (1, 8192) takes 17.04 s
|
| 93 |
+
(EngineCore pid=14912) INFO 11-19 00:10:58 [backends.py:916] collected artifacts: 65 entries, 20 artifacts, 24844810 bytes total
|
| 94 |
+
(EngineCore pid=14912) INFO 11-19 00:10:58 [decorators.py:717] saved AOT compiled function to /workspace/emc/work/cache/vllm/torch_compile_cache/torch_aot_compile/8a0bc292c67e43045b0fd3aa8597dbd95f65c054bf8675a0816e11306b67b354/rank_0_0/model
|
| 95 |
+
(EngineCore pid=14912) INFO 11-19 00:10:58 [monitor.py:53] torch.compile took 27.30 s in total
|
| 96 |
+
(EngineCore pid=14912) INFO 11-19 00:11:10 [monitor.py:81] Initial profiling/warmup run took 12.73 s
|
| 97 |
+
(EngineCore pid=14912) INFO 11-19 00:11:11 [backends.py:1090] Using cache directory: /workspace/emc/work/cache/vllm/torch_compile_cache/070dc5d8c5/rank_0_0/eagle_head for vLLM's torch.compile
|
| 98 |
+
(EngineCore pid=14912) INFO 11-19 00:11:11 [backends.py:1151] Dynamo bytecode transform time: 0.37 s
|
| 99 |
+
(EngineCore pid=14912) INFO 11-19 00:11:14 [backends.py:391] Compiling a graph for compile range (1, 8192) takes 3.04 s
|
| 100 |
+
(EngineCore pid=14912) INFO 11-19 00:11:14 [backends.py:916] collected artifacts: 2 entries, 2 artifacts, 3497242 bytes total
|
| 101 |
+
(EngineCore pid=14912) INFO 11-19 00:11:14 [decorators.py:717] saved AOT compiled function to /workspace/emc/work/cache/vllm/torch_compile_cache/torch_aot_compile/84189ba671cc1bd1f561bbaa20615ae8afd1e53a54fbbd7848261ae7a96d6fbe/rank_0_0/model
|
| 102 |
+
(EngineCore pid=14912) INFO 11-19 00:11:14 [monitor.py:53] torch.compile took 3.58 s in total
|
| 103 |
+
(EngineCore pid=14912) INFO 11-19 00:11:14 [monitor.py:81] Initial profiling/warmup run took 0.00 s
|
| 104 |
+
(EngineCore pid=14912) WARNING 11-19 00:11:16 [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 [00:36<03:37, 36.29s/it]
|
| 108 |
+
Capturing CUDA graphs (PIECEWISE): 29%|██▊ | 2/7 [00:36<01:15, 15.15s/it]
|
| 109 |
+
Capturing CUDA graphs (PIECEWISE): 43%|████▎ | 3/7 [00:36<00:33, 8.29s/it]
|
| 110 |
+
Capturing CUDA graphs (PIECEWISE): 57%|█████▋ | 4/7 [00:37<00:15, 5.13s/it]
|
| 111 |
+
Capturing CUDA graphs (PIECEWISE): 71%|███████▏ | 5/7 [01:30<00:45, 22.54s/it]
|
| 112 |
+
Capturing CUDA graphs (PIECEWISE): 100%|██████████| 7/7 [01:31<00:00, 11.33s/it]
|
| 113 |
+
Capturing CUDA graphs (PIECEWISE): 100%|██████████| 7/7 [01:31<00:00, 13.05s/it]
|
| 114 |
+
|
| 115 |
+
Capturing CUDA graphs (FULL): 0%| | 0/2 [00:00<?, ?it/s]
|
| 116 |
+
Capturing CUDA graphs (FULL): 50%|█████ | 1/2 [00:00<00:00, 4.17it/s]
|
| 117 |
+
Capturing CUDA graphs (FULL): 100%|██████████| 2/2 [00:00<00:00, 6.65it/s]
|
| 118 |
+
(EngineCore pid=14912) INFO 11-19 00:12:48 [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, 152.83it/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, 180.11it/s]
|
| 125 |
+
|
| 126 |
+
Capturing decode CUDA graphs (FULL): 0%| | 0/3 [00:00<?, ?it/s]
|
| 127 |
+
Capturing decode CUDA graphs (FULL): 33%|███▎ | 1/3 [00:00<00:00, 8.21it/s]
|
| 128 |
+
Capturing decode CUDA graphs (FULL): 100%|██████████| 3/3 [00:00<00:00, 21.49it/s]
|
| 129 |
+
(EngineCore pid=14912) INFO 11-19 00:12:48 [model_runner.py:1087] Graph capturing finished in 93 secs, took 0.23 GiB
|
| 130 |
+
(EngineCore pid=14912) INFO 11-19 00:12:49 [gpu_worker.py:692] Available KV cache memory: 30.91 GiB
|
| 131 |
+
(EngineCore pid=14912) INFO 11-19 00:12:49 [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.
|
| 132 |
+
(EngineCore pid=14912) WARNING 11-19 00:12:49 [kv_cache_utils.py:1574] Add 3 padding layers, may waste at most 6.25% KV cache memory
|
| 133 |
+
(EngineCore pid=14912) INFO 11-19 00:12:49 [kv_cache_utils.py:2464] GPU KV cache size: 295,408 tokens, Maximum concurrency for 16,384 tokens per request: 18.03x
|
| 134 |
+
(EngineCore pid=14912) INFO 11-19 00:12:49 [kernel_warmup.py:172] JIT kernel warmup starting.
|
| 135 |
+
(EngineCore pid=14912) INFO 11-19 00:12:49 [kernel_warmup.py:185] JIT kernel warmup finished in 0.00s.
|
| 136 |
+
(EngineCore pid=14912) INFO 11-19 00:12:49 [qwen_triton_warmup.py:294] Warming up Qwen GDN Triton kernels for model_type=qwen3_5_text.
|
| 137 |
+
(EngineCore pid=14912) INFO 11-19 00:12:49 [qwen_vl_triton_warmup.py:98] Warmed M-RoPE Triton kernels.
|
| 138 |
+
(EngineCore pid=14912) INFO 11-19 00:12:49 [mamba_triton_warmup.py:42] Warmed Mamba batch_memcpy_kernel.
|
| 139 |
+
(EngineCore pid=14912) INFO 11-19 00:12:50 [kernel_warmup.py:437] Using FlashInfer autotune cache file: /workspace/emc/work/cache/vllm/flashinfer_autotune_cache/0.7.0.post1/120f/0881ec9c562a209723a6b6cf4fcfccee53df672972c395bfe03858f09816333d/autotune_configs.json
|
| 140 |
+
2026-11-19 00:12:50,795 - INFO - autotuner.py:1046 - flashinfer.jit: [Autotuner]: Autotuning process starts ...
|
| 141 |
+
(EngineCore pid=14912) INFO 11-19 00:12:50 [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).
|
| 142 |
+
2026-11-19 00:12:50,817 - INFO - autotuner.py:1069 - flashinfer.jit: [Autotuner]: Autotuning process ends
|
| 143 |
+
2026-11-19 00:12:50,969 - INFO - autotuner.py:3554 - flashinfer.jit: [Autotuner]: Saved 0 configs to /workspace/emc/work/cache/vllm/flashinfer_autotune_cache/0.7.0.post1/120f/0881ec9c562a209723a6b6cf4fcfccee53df672972c395bfe03858f09816333d/autotune_configs.json (0 new, 0 from previous config)
|
| 144 |
+
|
| 145 |
+
Capturing CUDA graphs (PIECEWISE): 0%| | 0/7 [00:00<?, ?it/s]
|
| 146 |
+
Capturing CUDA graphs (PIECEWISE): 14%|█▍ | 1/7 [00:00<00:00, 7.05it/s]
|
| 147 |
+
Capturing CUDA graphs (PIECEWISE): 29%|██▊ | 2/7 [00:00<00:00, 7.05it/s]
|
| 148 |
+
Capturing CUDA graphs (PIECEWISE): 43%|████▎ | 3/7 [00:00<00:00, 7.18it/s]
|
| 149 |
+
Capturing CUDA graphs (PIECEWISE): 57%|█████▋ | 4/7 [00:00<00:00, 7.23it/s]
|
| 150 |
+
Capturing CUDA graphs (PIECEWISE): 86%|████████▌ | 6/7 [00:00<00:00, 8.99it/s]
|
| 151 |
+
Capturing CUDA graphs (PIECEWISE): 100%|██████████| 7/7 [00:00<00:00, 8.59it/s]
|
| 152 |
+
|
| 153 |
+
Capturing CUDA graphs (FULL): 0%| | 0/3 [00:00<?, ?it/s]
|
| 154 |
+
Capturing CUDA graphs (FULL): 67%|██████▋ | 2/3 [00:00<00:00, 16.33it/s]
|
| 155 |
+
Capturing CUDA graphs (FULL): 100%|██████████| 3/3 [00:00<00:00, 16.35it/s]
|
| 156 |
+
(EngineCore pid=14912) INFO 11-19 00:13:09 [speculator.py:144] Capturing model for speculator...
|
| 157 |
+
|
| 158 |
+
Capturing prefill CUDA graphs (PIECEWISE): 0%| | 0/7 [00:00<?, ?it/s]
|
| 159 |
+
Capturing prefill CUDA graphs (PIECEWISE): 100%|██████████| 7/7 [00:00<00:00, 150.94it/s]
|
| 160 |
+
|
| 161 |
+
Capturing prefill CUDA graphs (FULL): 0%| | 0/3 [00:00<?, ?it/s]
|
| 162 |
+
Capturing prefill CUDA graphs (FULL): 100%|██████████| 3/3 [00:00<00:00, 187.57it/s]
|
| 163 |
+
|
| 164 |
+
Capturing decode CUDA graphs (FULL): 0%| | 0/3 [00:00<?, ?it/s]
|
| 165 |
+
Capturing decode CUDA graphs (FULL): 100%|██████████| 3/3 [00:00<00:00, 111.79it/s]
|
| 166 |
+
(EngineCore pid=14912) INFO 11-19 00:13:09 [model_runner.py:1087] Graph capturing finished in 2 secs, took 0.17 GiB
|
| 167 |
+
(EngineCore pid=14912) INFO 11-19 00:13:09 [gpu_worker.py:892] CUDA graph pool memory: 0.17 GiB (actual), 0.24 GiB (estimated), difference: 0.07 GiB (40.7%).
|
| 168 |
+
(EngineCore pid=14912) INFO 11-19 00:13:09 [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.
|
| 169 |
+
(EngineCore pid=14912) INFO 11-19 00:13:10 [jit_monitor.py:84] Kernel JIT monitor activated; monitored JIT compilations during inference will use mode=warn.
|
| 170 |
+
(EngineCore pid=14912) INFO 11-19 00:13:11 [core.py:373] init engine (profile, create kv cache, warmup model) took 161.35 s (compilation: 30.88 s)
|
| 171 |
+
(EngineCore pid=14912) INFO 11-19 00:13:11 [kv_cache_utils.py:772] kv cache group sizes [16384, 16384, 16384, 800]
|
| 172 |
+
(EngineCore pid=14912) INFO 11-19 00:13:11 [kv_cache_utils.py:773] kv lcm block sizes 409600
|
| 173 |
+
(EngineCore pid=14912) INFO 11-19 00:13:11 [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'])
|
| 174 |
+
(APIServer pid=14373) INFO 11-19 00:13:11 [entry.py:132] Supported tasks: ['generate']
|
| 175 |
+
(APIServer pid=14373) INFO 11-19 00:13:11 [factories.py:76] Scale-out endpoints are disabled. Set --enable-scale-out to enable them.
|
| 176 |
+
(APIServer pid=14373) INFO 11-19 00:13:12 [hf.py:646] Detected the chat template content format to be 'openai'. You can set `--chat-template-content-format` to override this.
|
| 177 |
+
(APIServer pid=14373) WARNING 11-19 00:13:12 [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=14373) WARNING 11-19 00:13:12 [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.
|
| 179 |
+
(APIServer pid=14373) INFO 11-19 00:13:12 [entry.py:136] Starting vLLM server on http://127.0.0.1:53752
|
| 180 |
+
(APIServer pid=14373) INFO 11-19 00:13:12 [launcher.py:60] Available routes are:
|
| 181 |
+
(APIServer pid=14373) INFO 11-19 00:13:12 [launcher.py:69] Route: /openapi.json, Methods: GET, HEAD
|
| 182 |
+
(APIServer pid=14373) INFO 11-19 00:13:12 [launcher.py:69] Route: /docs, Methods: GET, HEAD
|
| 183 |
+
(APIServer pid=14373) INFO 11-19 00:13:12 [launcher.py:69] Route: /docs/oauth2-redirect, Methods: GET, HEAD
|
| 184 |
+
(APIServer pid=14373) INFO 11-19 00:13:12 [launcher.py:69] Route: /redoc, Methods: GET, HEAD
|
| 185 |
+
(APIServer pid=14373) INFO 11-19 00:13:12 [launcher.py:69] Route: /load, Methods: GET
|
| 186 |
+
(APIServer pid=14373) INFO 11-19 00:13:12 [launcher.py:69] Route: /version, Methods: GET
|
| 187 |
+
(APIServer pid=14373) INFO 11-19 00:13:12 [launcher.py:69] Route: /kv_event_sources, Methods: GET
|
| 188 |
+
(APIServer pid=14373) INFO 11-19 00:13:12 [launcher.py:69] Route: /health, Methods: GET
|
| 189 |
+
(APIServer pid=14373) INFO 11-19 00:13:12 [launcher.py:69] Route: /metrics, Methods: GET
|
| 190 |
+
(APIServer pid=14373) INFO 11-19 00:13:12 [launcher.py:69] Route: /tokenize, Methods: POST
|
| 191 |
+
(APIServer pid=14373) INFO 11-19 00:13:12 [launcher.py:69] Route: /detokenize, Methods: POST
|
| 192 |
+
(APIServer pid=14373) INFO 11-19 00:13:12 [launcher.py:69] Route: /v1/models, Methods: GET
|
| 193 |
+
(APIServer pid=14373) INFO 11-19 00:13:12 [launcher.py:69] Route: /ping, Methods: GET
|
| 194 |
+
(APIServer pid=14373) INFO 11-19 00:13:12 [launcher.py:69] Route: /ping, Methods: POST
|
| 195 |
+
(APIServer pid=14373) INFO 11-19 00:13:12 [launcher.py:69] Route: /invocations, Methods: POST
|
| 196 |
+
(APIServer pid=14373) INFO 11-19 00:13:12 [launcher.py:69] Route: /v1/chat/completions, Methods: POST
|
| 197 |
+
(APIServer pid=14373) INFO 11-19 00:13:12 [launcher.py:69] Route: /v1/chat/completions/batch, Methods: POST
|
| 198 |
+
(APIServer pid=14373) INFO 11-19 00:13:12 [launcher.py:69] Route: /v1/responses, Methods: POST
|
| 199 |
+
(APIServer pid=14373) INFO 11-19 00:13:12 [launcher.py:69] Route: /v1/responses/{response_id}, Methods: GET
|
| 200 |
+
(APIServer pid=14373) INFO 11-19 00:13:12 [launcher.py:69] Route: /v1/responses/{response_id}/cancel, Methods: POST
|
| 201 |
+
(APIServer pid=14373) INFO 11-19 00:13:12 [launcher.py:69] Route: /v1/completions, Methods: POST
|
| 202 |
+
(APIServer pid=14373) INFO 11-19 00:13:12 [launcher.py:69] Route: /v1/messages, Methods: POST
|
| 203 |
+
(APIServer pid=14373) INFO 11-19 00:13:12 [launcher.py:69] Route: /v1/messages/count_tokens, Methods: POST
|
| 204 |
+
(APIServer pid=14373) INFO 11-19 00:13:12 [launcher.py:69] Route: /generative_scoring, Methods: POST
|
| 205 |
+
(APIServer pid=14373) INFO 11-19 00:13:12 [launcher.py:69] Route: /scale_elastic_ep, Methods: POST
|
| 206 |
+
(APIServer pid=14373) INFO 11-19 00:13:12 [launcher.py:69] Route: /is_scaling_elastic_ep, Methods: POST
|
| 207 |
+
INFO: Started server process [14373]
|
| 208 |
+
INFO: Waiting for application startup.
|
| 209 |
+
INFO: Application startup complete.
|
| 210 |
+
INFO: 127.0.0.1:64866 - "GET /health HTTP/1.1" 200 OK
|
| 211 |
+
INFO: 127.0.0.1:18913 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 212 |
+
(EngineCore pid=14912) WARNING 11-19 00:13:14 [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.
|
| 213 |
+
(EngineCore pid=14912) WARNING 11-19 00:13:14 [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.
|
| 214 |
+
(EngineCore pid=14912) WARNING 11-19 00:13:14 [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.
|
| 215 |
+
INFO: 127.0.0.1:43472 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 216 |
+
(APIServer pid=14373) INFO 11-19 00:13:22 [loggers.py:320] Engine 000: Avg prompt throughput: 16.5 tokens/s, Avg generation throughput: 41.2 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 217 |
+
(APIServer pid=14373) INFO 11-19 00:13:22 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.15, Accepted throughput: 28.15 tokens/s, Drafted throughput: 39.30 tokens/s, Accepted: 318 tokens, Drafted: 444 tokens, Per-position acceptance rate: 0.824, 0.736, 0.588, Avg Draft acceptance rate: 71.6%
|
| 218 |
+
INFO: 127.0.0.1:20546 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 219 |
+
INFO: 127.0.0.1:17500 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 220 |
+
INFO: 127.0.0.1:47857 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 221 |
+
(APIServer pid=14373) INFO 11-19 00:13:32 [loggers.py:320] Engine 000: Avg prompt throughput: 42.7 tokens/s, Avg generation throughput: 63.1 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 222 |
+
(APIServer pid=14373) INFO 11-19 00:13:32 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.58, Accepted throughput: 45.60 tokens/s, Drafted throughput: 53.09 tokens/s, Accepted: 456 tokens, Drafted: 531 tokens, Per-position acceptance rate: 0.949, 0.859, 0.768, Avg Draft acceptance rate: 85.9%
|
| 223 |
+
INFO: 127.0.0.1:47729 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 224 |
+
INFO: 127.0.0.1:46690 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 225 |
+
(APIServer pid=14373) INFO 11-19 00:13:42 [loggers.py:320] Engine 000: Avg prompt throughput: 14.0 tokens/s, Avg generation throughput: 52.2 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 226 |
+
(APIServer pid=14373) INFO 11-19 00:13:42 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.88, Accepted throughput: 34.10 tokens/s, Drafted throughput: 54.29 tokens/s, Accepted: 341 tokens, Drafted: 543 tokens, Per-position acceptance rate: 0.829, 0.608, 0.448, Avg Draft acceptance rate: 62.8%
|
| 227 |
+
(APIServer pid=14373) INFO 11-19 00:13:52 [loggers.py:320] Engine 000: Avg prompt throughput: 0.0 tokens/s, Avg generation throughput: 43.3 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 228 |
+
(APIServer pid=14373) INFO 11-19 00:13:52 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.35, Accepted throughput: 24.90 tokens/s, Drafted throughput: 55.20 tokens/s, Accepted: 249 tokens, Drafted: 552 tokens, Per-position acceptance rate: 0.755, 0.386, 0.212, Avg Draft acceptance rate: 45.1%
|
| 229 |
+
INFO: 127.0.0.1:51753 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 230 |
+
(APIServer pid=14373) INFO 11-19 00:14:02 [loggers.py:320] Engine 000: Avg prompt throughput: 7.8 tokens/s, Avg generation throughput: 43.5 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 231 |
+
(APIServer pid=14373) INFO 11-19 00:14:02 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.40, Accepted throughput: 25.40 tokens/s, Drafted throughput: 54.60 tokens/s, Accepted: 254 tokens, Drafted: 546 tokens, Per-position acceptance rate: 0.753, 0.401, 0.242, Avg Draft acceptance rate: 46.5%
|
| 232 |
+
(APIServer pid=14373) INFO 11-19 00:14:12 [loggers.py:320] Engine 000: Avg prompt throughput: 0.0 tokens/s, Avg generation throughput: 41.1 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 233 |
+
(APIServer pid=14373) INFO 11-19 00:14:12 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.22, Accepted throughput: 22.60 tokens/s, Drafted throughput: 55.49 tokens/s, Accepted: 226 tokens, Drafted: 555 tokens, Per-position acceptance rate: 0.676, 0.357, 0.189, Avg Draft acceptance rate: 40.7%
|
| 234 |
+
(APIServer pid=14373) INFO 11-19 00:14:22 [loggers.py:320] Engine 000: Avg prompt throughput: 0.0 tokens/s, Avg generation throughput: 46.2 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.4%
|
| 235 |
+
(APIServer pid=14373) INFO 11-19 00:14:22 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.51, Accepted throughput: 27.80 tokens/s, Drafted throughput: 55.19 tokens/s, Accepted: 278 tokens, Drafted: 552 tokens, Per-position acceptance rate: 0.745, 0.473, 0.293, Avg Draft acceptance rate: 50.4%
|
| 236 |
+
INFO: 127.0.0.1:31863 - "GET /metrics HTTP/1.1" 200 OK
|
| 237 |
+
INFO: 127.0.0.1:36875 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 238 |
+
(APIServer pid=14373) INFO 11-19 00:14:32 [loggers.py:320] Engine 000: Avg prompt throughput: 7.2 tokens/s, Avg generation throughput: 48.0 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 239 |
+
(APIServer pid=14373) INFO 11-19 00:14:32 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.62, Accepted throughput: 29.60 tokens/s, Drafted throughput: 54.90 tokens/s, Accepted: 296 tokens, Drafted: 549 tokens, Per-position acceptance rate: 0.776, 0.519, 0.322, Avg Draft acceptance rate: 53.9%
|
| 240 |
+
INFO: 127.0.0.1:25686 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 241 |
+
(APIServer pid=14373) INFO 11-19 00:14:42 [loggers.py:320] Engine 000: Avg prompt throughput: 3.9 tokens/s, Avg generation throughput: 60.9 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
|
| 242 |
+
(APIServer pid=14373) INFO 11-19 00:14:42 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.32, Accepted throughput: 42.50 tokens/s, Drafted throughput: 54.90 tokens/s, Accepted: 425 tokens, Drafted: 549 tokens, Per-position acceptance rate: 0.880, 0.776, 0.667, Avg Draft acceptance rate: 77.4%
|
| 243 |
+
INFO: 127.0.0.1:31314 - "GET /metrics HTTP/1.1" 200 OK
|
| 244 |
+
INFO: 127.0.0.1:19036 - "GET /metrics HTTP/1.1" 200 OK
|
| 245 |
+
INFO: 127.0.0.1:51504 - "POST /v1/completions HTTP/1.1" 200 OK
|
| 246 |
+
INFO: 127.0.0.1:50818 - "POST /v1/completions HTTP/1.1" 200 OK
|
runs/r3/logs/pipeline.log
CHANGED
|
@@ -11,3 +11,4 @@
|
|
| 11 |
[2026-11-19 00:07:57] start probe
|
| 12 |
[2026-11-19 00:08:21] done probe (24s)
|
| 13 |
[2026-11-19 00:08:42] done prompts (56s)
|
|
|
|
|
|
| 11 |
[2026-11-19 00:07:57] start probe
|
| 12 |
[2026-11-19 00:08:21] done probe (24s)
|
| 13 |
[2026-11-19 00:08:42] done prompts (56s)
|
| 14 |
+
[2026-11-19 00:08:51] start smoke_fp8
|
runs/r3/logs/pipeline.out
CHANGED
|
@@ -11,3 +11,4 @@
|
|
| 11 |
[2026-11-19 00:07:57] start probe
|
| 12 |
[2026-11-19 00:08:21] done probe (24s)
|
| 13 |
[2026-11-19 00:08:42] done prompts (56s)
|
|
|
|
|
|
| 11 |
[2026-11-19 00:07:57] start probe
|
| 12 |
[2026-11-19 00:08:21] done probe (24s)
|
| 13 |
[2026-11-19 00:08:42] done prompts (56s)
|
| 14 |
+
[2026-11-19 00:08:51] start smoke_fp8
|
runs/r3/logs/smoke_fp8.log
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
method: mtp
|
| 2 |
+
parameters: 424.7 M
|
| 3 |
+
weights: 0.540 GB
|
| 4 |
+
KV cache @ full context: 0.067 GB
|
| 5 |
+
memory metric: 0.607 GB
|
| 6 |
+
PASS
|
| 7 |
+
[00:08:59] repeat 1/1: fresh vLLM process
|
| 8 |
+
[00:08:59] starting vLLM (log: /workspace/emc/work/logs/20261119-000859-eval-r1.log)
|
| 9 |
+
[00:13:13] vLLM ready after 254s
|
| 10 |
+
[00:14:42] chat: 2 prompts done
|
runs/r3/logs/stats.log
CHANGED
|
@@ -9,3 +9,5 @@
|
|
| 9 |
[2026-11-19 00:07:58] ram 1.5/172 GB used (cache 0.0, shmem 0.0) | cgroup 1.5/172 GB | disk 65 GB used | gpu 0.0/95.6 GiB, 0%, 79 W | load 0.0 | largest python 2.1 GB, python 0.1 GB, python 0.1 GB
|
| 10 |
[2026-11-19 00:08:21] ram 3.8/172 GB used (cache 0.0, shmem 0.0) | cgroup 3.8/172 GB | disk 66 GB used | gpu 0.0/95.6 GiB, 0%, 59 W | load 0.0 | largest python 1.7 GB, python 0.1 GB, python 0.1 GB
|
| 11 |
[2026-11-19 00:08:42] ram 0.2/172 GB used (cache 0.0, shmem 0.0) | cgroup 0.2/172 GB | disk 66 GB used | gpu 0.0/95.6 GiB, 0%, 59 W | load 0.0 | largest python 0.1 GB, python 0.1 GB, python 0.1 GB
|
|
|
|
|
|
|
|
|
| 9 |
[2026-11-19 00:07:58] ram 1.5/172 GB used (cache 0.0, shmem 0.0) | cgroup 1.5/172 GB | disk 65 GB used | gpu 0.0/95.6 GiB, 0%, 79 W | load 0.0 | largest python 2.1 GB, python 0.1 GB, python 0.1 GB
|
| 10 |
[2026-11-19 00:08:21] ram 3.8/172 GB used (cache 0.0, shmem 0.0) | cgroup 3.8/172 GB | disk 66 GB used | gpu 0.0/95.6 GiB, 0%, 59 W | load 0.0 | largest python 1.7 GB, python 0.1 GB, python 0.1 GB
|
| 11 |
[2026-11-19 00:08:42] ram 0.2/172 GB used (cache 0.0, shmem 0.0) | cgroup 0.2/172 GB | disk 66 GB used | gpu 0.0/95.6 GiB, 0%, 59 W | load 0.0 | largest python 0.1 GB, python 0.1 GB, python 0.1 GB
|
| 12 |
+
[2026-11-19 00:08:51] ram 0.2/172 GB used (cache 0.0, shmem 0.0) | cgroup 0.2/172 GB | disk 66 GB used | gpu 0.0/95.6 GiB, 0%, 59 W | load 0.0 | largest python 0.1 GB, python 0.1 GB, python 0.1 GB
|
| 13 |
+
[2026-11-19 00:14:47] ram 5.0/172 GB used (cache 0.3, shmem 0.0) | cgroup 5.0/172 GB | disk 66 GB used | gpu 83.1/95.6 GiB, 100%, 288 W | load 0.0 | largest VLLM::EngineCor 4.7 GB, vllm 2.5 GB, python 1.2 GB
|
runs/r3/logs/sync.log
CHANGED
|
@@ -3,3 +3,4 @@ uploaded logs/results to mudassiruddin21/EMC:runs/r3
|
|
| 3 |
uploaded logs/results to mudassiruddin21/EMC:runs/r3
|
| 4 |
uploaded logs/results to mudassiruddin21/EMC:runs/r3
|
| 5 |
uploaded logs/results to mudassiruddin21/EMC:runs/r3
|
|
|
|
|
|
| 3 |
uploaded logs/results to mudassiruddin21/EMC:runs/r3
|
| 4 |
uploaded logs/results to mudassiruddin21/EMC:runs/r3
|
| 5 |
uploaded logs/results to mudassiruddin21/EMC:runs/r3
|
| 6 |
+
uploaded logs/results to mudassiruddin21/EMC:runs/r3
|