2025-08-21 00:49:07 - INFO - Loading model: Qwen/Qwen2.5-VL-3B-Instruct-AWQ
2025-08-21 00:49:11 - INFO - We will use 90% of the memory on device 0 for storing the model, and 10% for the buffer to avoid OOM. You can set `max_memory` in to a higher value to use more memory (at your own risk).
2025-08-21 00:49:19 - INFO - Model loaded in 11.79 seconds
2025-08-21 00:49:19 - INFO - GPU Memory Usage after model load: 3250.55 MB
2025-08-21 00:50:53 - INFO - [30fe7962-43c7-418e-9663-3cf53776c810] Received new video inference request. Prompt: 'Summarize the key observable events in this 1-minute convenience store video clip. Focus strictly on the physical actions and interactions of the people. Describe only what you can see', Video: '/mnt/data/xiuying/Code/local_deploy/video/Clips_60s/sample_part_001.mp4'
2025-08-21 00:50:53 - INFO - [30fe7962-43c7-418e-9663-3cf53776c810] Video saved to temporary file: temp_videos/30fe7962-43c7-418e-9663-3cf53776c810.mp4
2025-08-21 00:50:53 - INFO - [30fe7962-43c7-418e-9663-3cf53776c810] Extracting frames using method: uniform, rate/threshold: 30
2025-08-21 00:50:58 - INFO - [30fe7962-43c7-418e-9663-3cf53776c810] Extracted 30 frames successfully. Saving to temporary files...
2025-08-21 00:50:58 - INFO - [30fe7962-43c7-418e-9663-3cf53776c810] 30 frames saved to temp_videos/30fe7962-43c7-418e-9663-3cf53776c810
2025-08-21 00:50:58 - INFO - Prompt token length: 2306
2025-08-21 00:51:05 - INFO - Tokens per second: 11.82581726573877, Peak GPU memory MB: 5348.375
2025-08-21 00:51:05 - INFO - [30fe7962-43c7-418e-9663-3cf53776c810] Inference time: 11.48 seconds, CPU usage: 19.9%, CPU core utilization: [17.8, 19.9, 21.8, 20.2]
2025-08-21 00:51:05 - INFO - [30fe7962-43c7-418e-9663-3cf53776c810] Cleaned up temporary frame directory: temp_videos/30fe7962-43c7-418e-9663-3cf53776c810
2025-08-21 00:51:05 - INFO - [a3af8f29-02fa-49b6-bcb7-c671f274c93a] Received new video inference request. Prompt: 'Summarize the key observable events in this 1-minute convenience store video clip. Focus strictly on the physical actions and interactions of the people. Describe only what you can see', Video: '/mnt/data/xiuying/Code/local_deploy/video/Clips_60s/sample_part_002.mp4'
2025-08-21 00:51:05 - INFO - [a3af8f29-02fa-49b6-bcb7-c671f274c93a] Video saved to temporary file: temp_videos/a3af8f29-02fa-49b6-bcb7-c671f274c93a.mp4
2025-08-21 00:51:05 - INFO - [a3af8f29-02fa-49b6-bcb7-c671f274c93a] Extracting frames using method: uniform, rate/threshold: 30
2025-08-21 00:51:09 - INFO - [a3af8f29-02fa-49b6-bcb7-c671f274c93a] Extracted 30 frames successfully. Saving to temporary files...
2025-08-21 00:51:09 - INFO - [a3af8f29-02fa-49b6-bcb7-c671f274c93a] 30 frames saved to temp_videos/a3af8f29-02fa-49b6-bcb7-c671f274c93a
2025-08-21 00:51:10 - INFO - Prompt token length: 2306
2025-08-21 00:51:15 - INFO - Tokens per second: 11.74072921403584, Peak GPU memory MB: 5348.375
2025-08-21 00:51:15 - INFO - [a3af8f29-02fa-49b6-bcb7-c671f274c93a] Inference time: 10.66 seconds, CPU usage: 56.2%, CPU core utilization: [43.8, 43.4, 43.8, 93.4]
2025-08-21 00:51:15 - INFO - [a3af8f29-02fa-49b6-bcb7-c671f274c93a] Cleaned up temporary frame directory: temp_videos/a3af8f29-02fa-49b6-bcb7-c671f274c93a
2025-08-21 00:51:15 - INFO - [be2cf942-7b83-46a1-80f4-3341fc34fdda] Received new video inference request. Prompt: 'Summarize the key observable events in this 1-minute convenience store video clip. Focus strictly on the physical actions and interactions of the people. Describe only what you can see', Video: '/mnt/data/xiuying/Code/local_deploy/video/Clips_60s/sample_part_003.mp4'
2025-08-21 00:51:15 - INFO - [be2cf942-7b83-46a1-80f4-3341fc34fdda] Video saved to temporary file: temp_videos/be2cf942-7b83-46a1-80f4-3341fc34fdda.mp4
2025-08-21 00:51:15 - INFO - [be2cf942-7b83-46a1-80f4-3341fc34fdda] Extracting frames using method: uniform, rate/threshold: 30
2025-08-21 00:51:20 - INFO - [be2cf942-7b83-46a1-80f4-3341fc34fdda] Extracted 30 frames successfully. Saving to temporary files...
2025-08-21 00:51:20 - INFO - [be2cf942-7b83-46a1-80f4-3341fc34fdda] 30 frames saved to temp_videos/be2cf942-7b83-46a1-80f4-3341fc34fdda
2025-08-21 00:51:20 - INFO - Prompt token length: 2306
2025-08-21 00:51:27 - INFO - Tokens per second: 11.73304837389127, Peak GPU memory MB: 5348.375
2025-08-21 00:51:27 - INFO - [be2cf942-7b83-46a1-80f4-3341fc34fdda] Inference time: 11.55 seconds, CPU usage: 52.1%, CPU core utilization: [38.3, 93.8, 38.3, 38.0]
2025-08-21 00:51:27 - INFO - [be2cf942-7b83-46a1-80f4-3341fc34fdda] Cleaned up temporary frame directory: temp_videos/be2cf942-7b83-46a1-80f4-3341fc34fdda
2025-08-21 00:51:27 - INFO - [1d4fb530-0fc9-438f-a51a-cabce02b6cb7] Received new video inference request. Prompt: 'Summarize the key observable events in this 1-minute convenience store video clip. Focus strictly on the physical actions and interactions of the people. Describe only what you can see', Video: '/mnt/data/xiuying/Code/local_deploy/video/Clips_60s/sample_part_004.mp4'
2025-08-21 00:51:27 - INFO - [1d4fb530-0fc9-438f-a51a-cabce02b6cb7] Video saved to temporary file: temp_videos/1d4fb530-0fc9-438f-a51a-cabce02b6cb7.mp4
2025-08-21 00:51:27 - INFO - [1d4fb530-0fc9-438f-a51a-cabce02b6cb7] Extracting frames using method: uniform, rate/threshold: 30
2025-08-21 00:51:32 - INFO - [1d4fb530-0fc9-438f-a51a-cabce02b6cb7] Extracted 30 frames successfully. Saving to temporary files...
2025-08-21 00:51:32 - INFO - [1d4fb530-0fc9-438f-a51a-cabce02b6cb7] 30 frames saved to temp_videos/1d4fb530-0fc9-438f-a51a-cabce02b6cb7
2025-08-21 00:51:32 - INFO - Prompt token length: 2306
2025-08-21 00:51:38 - INFO - Tokens per second: 11.929480284506932, Peak GPU memory MB: 5348.375
2025-08-21 00:51:38 - INFO - [1d4fb530-0fc9-438f-a51a-cabce02b6cb7] Inference time: 11.57 seconds, CPU usage: 52.3%, CPU core utilization: [89.2, 38.6, 42.9, 38.3]
2025-08-21 00:51:38 - INFO - [1d4fb530-0fc9-438f-a51a-cabce02b6cb7] Cleaned up temporary frame directory: temp_videos/1d4fb530-0fc9-438f-a51a-cabce02b6cb7
2025-08-21 00:51:38 - INFO - [ce11e297-0569-49ac-85bc-050e43e84448] Received new video inference request. Prompt: 'Summarize the key observable events in this 1-minute convenience store video clip. Focus strictly on the physical actions and interactions of the people. Describe only what you can see', Video: '/mnt/data/xiuying/Code/local_deploy/video/Clips_60s/sample_part_005.mp4'
2025-08-21 00:51:38 - INFO - [ce11e297-0569-49ac-85bc-050e43e84448] Video saved to temporary file: temp_videos/ce11e297-0569-49ac-85bc-050e43e84448.mp4
2025-08-21 00:51:38 - INFO - [ce11e297-0569-49ac-85bc-050e43e84448] Extracting frames using method: uniform, rate/threshold: 30
2025-08-21 00:51:43 - INFO - [ce11e297-0569-49ac-85bc-050e43e84448] Extracted 30 frames successfully. Saving to temporary files...
2025-08-21 00:51:43 - INFO - [ce11e297-0569-49ac-85bc-050e43e84448] 30 frames saved to temp_videos/ce11e297-0569-49ac-85bc-050e43e84448
2025-08-21 00:51:43 - INFO - Prompt token length: 2306
2025-08-21 00:51:50 - INFO - Tokens per second: 11.89941941740383, Peak GPU memory MB: 5348.375
2025-08-21 00:51:50 - INFO - [ce11e297-0569-49ac-85bc-050e43e84448] Inference time: 11.12 seconds, CPU usage: 53.2%, CPU core utilization: [37.8, 40.6, 93.6, 40.6]
2025-08-21 00:51:50 - INFO - [ce11e297-0569-49ac-85bc-050e43e84448] Cleaned up temporary frame directory: temp_videos/ce11e297-0569-49ac-85bc-050e43e84448
2025-08-21 00:51:50 - INFO - [5cec6dd5-3430-473f-aa6a-0d81b6475f34] Received new video inference request. Prompt: 'Summarize the key observable events in this 1-minute convenience store video clip. Focus strictly on the physical actions and interactions of the people. Describe only what you can see', Video: '/mnt/data/xiuying/Code/local_deploy/video/Clips_60s/sample_part_006.mp4'
2025-08-21 00:51:50 - INFO - [5cec6dd5-3430-473f-aa6a-0d81b6475f34] Video saved to temporary file: temp_videos/5cec6dd5-3430-473f-aa6a-0d81b6475f34.mp4
2025-08-21 00:51:50 - INFO - [5cec6dd5-3430-473f-aa6a-0d81b6475f34] Extracting frames using method: uniform, rate/threshold: 30
2025-08-21 00:51:54 - INFO - [5cec6dd5-3430-473f-aa6a-0d81b6475f34] Extracted 30 frames successfully. Saving to temporary files...
2025-08-21 00:51:54 - INFO - [5cec6dd5-3430-473f-aa6a-0d81b6475f34] 30 frames saved to temp_videos/5cec6dd5-3430-473f-aa6a-0d81b6475f34
2025-08-21 00:51:55 - INFO - Prompt token length: 2306
2025-08-21 00:52:00 - INFO - Tokens per second: 11.881699260124632, Peak GPU memory MB: 5348.375
2025-08-21 00:52:00 - INFO - [5cec6dd5-3430-473f-aa6a-0d81b6475f34] Inference time: 10.77 seconds, CPU usage: 53.6%, CPU core utilization: [59.2, 66.8, 47.9, 40.4]
2025-08-21 00:52:00 - INFO - [5cec6dd5-3430-473f-aa6a-0d81b6475f34] Cleaned up temporary frame directory: temp_videos/5cec6dd5-3430-473f-aa6a-0d81b6475f34
2025-08-21 00:52:21 - INFO - [70289040-01c3-4ed8-83de-5a2d9996ed2d] Received new video inference request. Prompt: 'Summarize the key events in this convenience store video. Focus only on the actions and interactions of the people. Avoid repetitive descriptions of the store's layout or shelves.', Video: '/mnt/data/xiuying/Code/local_deploy/video/Clips_60s/sample_part_001.mp4'
2025-08-21 00:52:21 - INFO - [70289040-01c3-4ed8-83de-5a2d9996ed2d] Video saved to temporary file: temp_videos/70289040-01c3-4ed8-83de-5a2d9996ed2d.mp4
2025-08-21 00:52:21 - INFO - [70289040-01c3-4ed8-83de-5a2d9996ed2d] Extracting frames using method: uniform, rate/threshold: 30
2025-08-21 00:52:26 - INFO - [70289040-01c3-4ed8-83de-5a2d9996ed2d] Extracted 30 frames successfully. Saving to temporary files...
2025-08-21 00:52:26 - INFO - [70289040-01c3-4ed8-83de-5a2d9996ed2d] 30 frames saved to temp_videos/70289040-01c3-4ed8-83de-5a2d9996ed2d
2025-08-21 00:52:26 - INFO - Prompt token length: 2305
2025-08-21 00:52:32 - INFO - Tokens per second: 11.87194334609885, Peak GPU memory MB: 5348.375
2025-08-21 00:52:32 - INFO - [70289040-01c3-4ed8-83de-5a2d9996ed2d] Inference time: 10.96 seconds, CPU usage: 20.3%, CPU core utilization: [15.5, 15.3, 33.8, 16.7]
2025-08-21 00:52:32 - INFO - [70289040-01c3-4ed8-83de-5a2d9996ed2d] Cleaned up temporary frame directory: temp_videos/70289040-01c3-4ed8-83de-5a2d9996ed2d
2025-08-21 00:52:32 - INFO - [a8bb150b-138f-4300-adf1-fa15dbace647] Received new video inference request. Prompt: 'Summarize the key events in this convenience store video. Focus only on the actions and interactions of the people. Avoid repetitive descriptions of the store's layout or shelves.', Video: '/mnt/data/xiuying/Code/local_deploy/video/Clips_60s/sample_part_002.mp4'
2025-08-21 00:52:32 - INFO - [a8bb150b-138f-4300-adf1-fa15dbace647] Video saved to temporary file: temp_videos/a8bb150b-138f-4300-adf1-fa15dbace647.mp4
2025-08-21 00:52:32 - INFO - [a8bb150b-138f-4300-adf1-fa15dbace647] Extracting frames using method: uniform, rate/threshold: 30
2025-08-21 00:52:37 - INFO - [a8bb150b-138f-4300-adf1-fa15dbace647] Extracted 30 frames successfully. Saving to temporary files...
2025-08-21 00:52:37 - INFO - [a8bb150b-138f-4300-adf1-fa15dbace647] 30 frames saved to temp_videos/a8bb150b-138f-4300-adf1-fa15dbace647
2025-08-21 00:52:37 - INFO - Prompt token length: 2305
2025-08-21 00:52:44 - INFO - Tokens per second: 11.83286096302887, Peak GPU memory MB: 5348.375
2025-08-21 00:52:44 - INFO - [a8bb150b-138f-4300-adf1-fa15dbace647] Inference time: 11.96 seconds, CPU usage: 52.1%, CPU core utilization: [39.5, 38.2, 37.0, 93.5]
2025-08-21 00:52:44 - INFO - [a8bb150b-138f-4300-adf1-fa15dbace647] Cleaned up temporary frame directory: temp_videos/a8bb150b-138f-4300-adf1-fa15dbace647
2025-08-21 00:52:44 - INFO - [6f5c9723-cae6-47ab-8cc7-6942f1ad38d4] Received new video inference request. Prompt: 'Summarize the key events in this convenience store video. Focus only on the actions and interactions of the people. Avoid repetitive descriptions of the store's layout or shelves.', Video: '/mnt/data/xiuying/Code/local_deploy/video/Clips_60s/sample_part_003.mp4'
2025-08-21 00:52:44 - INFO - [6f5c9723-cae6-47ab-8cc7-6942f1ad38d4] Video saved to temporary file: temp_videos/6f5c9723-cae6-47ab-8cc7-6942f1ad38d4.mp4
2025-08-21 00:52:44 - INFO - [6f5c9723-cae6-47ab-8cc7-6942f1ad38d4] Extracting frames using method: uniform, rate/threshold: 30
2025-08-21 00:52:49 - INFO - [6f5c9723-cae6-47ab-8cc7-6942f1ad38d4] Extracted 30 frames successfully. Saving to temporary files...
2025-08-21 00:52:49 - INFO - [6f5c9723-cae6-47ab-8cc7-6942f1ad38d4] 30 frames saved to temp_videos/6f5c9723-cae6-47ab-8cc7-6942f1ad38d4
2025-08-21 00:52:49 - INFO - Prompt token length: 2305
2025-08-21 00:52:56 - INFO - Tokens per second: 11.760994323642526, Peak GPU memory MB: 5348.375
2025-08-21 00:52:56 - INFO - [6f5c9723-cae6-47ab-8cc7-6942f1ad38d4] Inference time: 11.29 seconds, CPU usage: 53.3%, CPU core utilization: [40.3, 75.1, 39.1, 58.8]
2025-08-21 00:52:56 - INFO - [6f5c9723-cae6-47ab-8cc7-6942f1ad38d4] Cleaned up temporary frame directory: temp_videos/6f5c9723-cae6-47ab-8cc7-6942f1ad38d4
2025-08-21 00:52:56 - INFO - [0d1e6fde-165d-4649-8878-c71f32a33f71] Received new video inference request. Prompt: 'Summarize the key events in this convenience store video. Focus only on the actions and interactions of the people. Avoid repetitive descriptions of the store's layout or shelves.', Video: '/mnt/data/xiuying/Code/local_deploy/video/Clips_60s/sample_part_004.mp4'
2025-08-21 00:52:56 - INFO - [0d1e6fde-165d-4649-8878-c71f32a33f71] Video saved to temporary file: temp_videos/0d1e6fde-165d-4649-8878-c71f32a33f71.mp4
2025-08-21 00:52:56 - INFO - [0d1e6fde-165d-4649-8878-c71f32a33f71] Extracting frames using method: uniform, rate/threshold: 30
2025-08-21 00:53:00 - INFO - [0d1e6fde-165d-4649-8878-c71f32a33f71] Extracted 30 frames successfully. Saving to temporary files...
2025-08-21 00:53:01 - INFO - [0d1e6fde-165d-4649-8878-c71f32a33f71] 30 frames saved to temp_videos/0d1e6fde-165d-4649-8878-c71f32a33f71
2025-08-21 00:53:01 - INFO - Prompt token length: 2305
2025-08-21 00:53:07 - INFO - Tokens per second: 11.888364051217732, Peak GPU memory MB: 5348.375
2025-08-21 00:53:07 - INFO - [0d1e6fde-165d-4649-8878-c71f32a33f71] Inference time: 11.70 seconds, CPU usage: 51.6%, CPU core utilization: [59.1, 67.7, 41.2, 38.5]
2025-08-21 00:53:07 - INFO - [0d1e6fde-165d-4649-8878-c71f32a33f71] Cleaned up temporary frame directory: temp_videos/0d1e6fde-165d-4649-8878-c71f32a33f71
2025-08-21 00:53:07 - INFO - [884171ad-eeda-4dd5-9f1b-d43868fa7804] Received new video inference request. Prompt: 'Summarize the key events in this convenience store video. Focus only on the actions and interactions of the people. Avoid repetitive descriptions of the store's layout or shelves.', Video: '/mnt/data/xiuying/Code/local_deploy/video/Clips_60s/sample_part_005.mp4'
2025-08-21 00:53:07 - INFO - [884171ad-eeda-4dd5-9f1b-d43868fa7804] Video saved to temporary file: temp_videos/884171ad-eeda-4dd5-9f1b-d43868fa7804.mp4
2025-08-21 00:53:07 - INFO - [884171ad-eeda-4dd5-9f1b-d43868fa7804] Extracting frames using method: uniform, rate/threshold: 30
2025-08-21 00:53:12 - INFO - [884171ad-eeda-4dd5-9f1b-d43868fa7804] Extracted 30 frames successfully. Saving to temporary files...
2025-08-21 00:53:12 - INFO - [884171ad-eeda-4dd5-9f1b-d43868fa7804] 30 frames saved to temp_videos/884171ad-eeda-4dd5-9f1b-d43868fa7804
2025-08-21 00:53:12 - INFO - Prompt token length: 2305
2025-08-21 00:53:18 - INFO - Tokens per second: 11.790592742669173, Peak GPU memory MB: 5348.375
2025-08-21 00:53:18 - INFO - [884171ad-eeda-4dd5-9f1b-d43868fa7804] Inference time: 10.74 seconds, CPU usage: 55.3%, CPU core utilization: [53.6, 43.0, 81.8, 42.7]
2025-08-21 00:53:18 - INFO - [884171ad-eeda-4dd5-9f1b-d43868fa7804] Cleaned up temporary frame directory: temp_videos/884171ad-eeda-4dd5-9f1b-d43868fa7804
2025-08-21 00:53:18 - INFO - [ce71041b-d894-486d-9e37-8b9a86705705] Received new video inference request. Prompt: 'Summarize the key events in this convenience store video. Focus only on the actions and interactions of the people. Avoid repetitive descriptions of the store's layout or shelves.', Video: '/mnt/data/xiuying/Code/local_deploy/video/Clips_60s/sample_part_006.mp4'
2025-08-21 00:53:18 - INFO - [ce71041b-d894-486d-9e37-8b9a86705705] Video saved to temporary file: temp_videos/ce71041b-d894-486d-9e37-8b9a86705705.mp4
2025-08-21 00:53:18 - INFO - [ce71041b-d894-486d-9e37-8b9a86705705] Extracting frames using method: uniform, rate/threshold: 30
2025-08-21 00:53:23 - INFO - [ce71041b-d894-486d-9e37-8b9a86705705] Extracted 30 frames successfully. Saving to temporary files...
2025-08-21 00:53:23 - INFO - [ce71041b-d894-486d-9e37-8b9a86705705] 30 frames saved to temp_videos/ce71041b-d894-486d-9e37-8b9a86705705
2025-08-21 00:53:23 - INFO - Prompt token length: 2305
2025-08-21 00:53:28 - INFO - Tokens per second: 11.854764118772119, Peak GPU memory MB: 5348.375
2025-08-21 00:53:28 - INFO - [ce71041b-d894-486d-9e37-8b9a86705705] Inference time: 10.06 seconds, CPU usage: 55.9%, CPU core utilization: [48.9, 44.4, 87.7, 42.4]
2025-08-21 00:53:28 - INFO - [ce71041b-d894-486d-9e37-8b9a86705705] Cleaned up temporary frame directory: temp_videos/ce71041b-d894-486d-9e37-8b9a86705705
2025-08-21 00:53:28 - INFO - [a135f93a-1ac9-4578-a1c4-2b1aeb89afda] Received new video inference request. Prompt: 'Summarize the key events in this convenience store video. Focus only on the actions and interactions of the people. Avoid repetitive descriptions of the store's layout or shelves.', Video: '/mnt/data/xiuying/Code/local_deploy/video/Clips_60s/sample_part_007.mp4'
2025-08-21 00:53:28 - INFO - [a135f93a-1ac9-4578-a1c4-2b1aeb89afda] Video saved to temporary file: temp_videos/a135f93a-1ac9-4578-a1c4-2b1aeb89afda.mp4
2025-08-21 00:53:28 - INFO - [a135f93a-1ac9-4578-a1c4-2b1aeb89afda] Extracting frames using method: uniform, rate/threshold: 30
2025-08-21 00:53:33 - INFO - [a135f93a-1ac9-4578-a1c4-2b1aeb89afda] Extracted 30 frames successfully. Saving to temporary files...
2025-08-21 00:53:33 - INFO - [a135f93a-1ac9-4578-a1c4-2b1aeb89afda] 30 frames saved to temp_videos/a135f93a-1ac9-4578-a1c4-2b1aeb89afda
2025-08-21 00:53:33 - INFO - Prompt token length: 2305
2025-08-21 00:53:40 - INFO - Tokens per second: 11.806017274209756, Peak GPU memory MB: 5348.375
2025-08-21 00:53:40 - INFO - [a135f93a-1ac9-4578-a1c4-2b1aeb89afda] Inference time: 12.00 seconds, CPU usage: 51.4%, CPU core utilization: [49.4, 37.7, 81.8, 36.8]
2025-08-21 00:53:40 - INFO - [a135f93a-1ac9-4578-a1c4-2b1aeb89afda] Cleaned up temporary frame directory: temp_videos/a135f93a-1ac9-4578-a1c4-2b1aeb89afda
2025-08-21 00:53:40 - INFO - [6873e58b-3473-4224-909d-3159c03588e5] Received new video inference request. Prompt: 'Summarize the key events in this convenience store video. Focus only on the actions and interactions of the people. Avoid repetitive descriptions of the store's layout or shelves.', Video: '/mnt/data/xiuying/Code/local_deploy/video/Clips_60s/sample_part_008.mp4'
2025-08-21 00:53:40 - INFO - [6873e58b-3473-4224-909d-3159c03588e5] Video saved to temporary file: temp_videos/6873e58b-3473-4224-909d-3159c03588e5.mp4
2025-08-21 00:53:40 - INFO - [6873e58b-3473-4224-909d-3159c03588e5] Extracting frames using method: uniform, rate/threshold: 30
2025-08-21 00:53:45 - INFO - [6873e58b-3473-4224-909d-3159c03588e5] Extracted 30 frames successfully. Saving to temporary files...
2025-08-21 00:53:45 - INFO - [6873e58b-3473-4224-909d-3159c03588e5] 30 frames saved to temp_videos/6873e58b-3473-4224-909d-3159c03588e5
2025-08-21 00:53:45 - INFO - Prompt token length: 2305
2025-08-21 00:53:51 - INFO - Tokens per second: 11.878890265234213, Peak GPU memory MB: 5348.375
2025-08-21 00:53:51 - INFO - [6873e58b-3473-4224-909d-3159c03588e5] Inference time: 10.61 seconds, CPU usage: 55.0%, CPU core utilization: [90.1, 42.2, 45.8, 41.8]
2025-08-21 00:53:51 - INFO - [6873e58b-3473-4224-909d-3159c03588e5] Cleaned up temporary frame directory: temp_videos/6873e58b-3473-4224-909d-3159c03588e5
2025-08-21 00:53:51 - INFO - [6356c394-1484-4391-b145-81215ba47ee8] Received new video inference request. Prompt: 'Summarize the key events in this convenience store video. Focus only on the actions and interactions of the people. Avoid repetitive descriptions of the store's layout or shelves.', Video: '/mnt/data/xiuying/Code/local_deploy/video/Clips_60s/sample_part_009.mp4'
2025-08-21 00:53:51 - INFO - [6356c394-1484-4391-b145-81215ba47ee8] Video saved to temporary file: temp_videos/6356c394-1484-4391-b145-81215ba47ee8.mp4
2025-08-21 00:53:51 - INFO - [6356c394-1484-4391-b145-81215ba47ee8] Extracting frames using method: uniform, rate/threshold: 30
2025-08-21 00:53:56 - INFO - [6356c394-1484-4391-b145-81215ba47ee8] Extracted 30 frames successfully. Saving to temporary files...
2025-08-21 00:53:56 - INFO - [6356c394-1484-4391-b145-81215ba47ee8] 30 frames saved to temp_videos/6356c394-1484-4391-b145-81215ba47ee8
2025-08-21 00:53:56 - INFO - Prompt token length: 2305
2025-08-21 00:54:02 - INFO - Tokens per second: 11.82995179235076, Peak GPU memory MB: 5348.375
2025-08-21 00:54:02 - INFO - [6356c394-1484-4391-b145-81215ba47ee8] Inference time: 10.80 seconds, CPU usage: 53.8%, CPU core utilization: [49.9, 41.3, 78.9, 45.0]
2025-08-21 00:54:02 - INFO - [6356c394-1484-4391-b145-81215ba47ee8] Cleaned up temporary frame directory: temp_videos/6356c394-1484-4391-b145-81215ba47ee8
2025-08-21 00:54:02 - INFO - [2999105b-10bc-497e-8931-352c7d9d65e6] Received new video inference request. Prompt: 'Summarize the key events in this convenience store video. Focus only on the actions and interactions of the people. Avoid repetitive descriptions of the store's layout or shelves.', Video: '/mnt/data/xiuying/Code/local_deploy/video/Clips_60s/sample_part_010.mp4'
2025-08-21 00:54:02 - INFO - [2999105b-10bc-497e-8931-352c7d9d65e6] Video saved to temporary file: temp_videos/2999105b-10bc-497e-8931-352c7d9d65e6.mp4
2025-08-21 00:54:02 - INFO - [2999105b-10bc-497e-8931-352c7d9d65e6] Extracting frames using method: uniform, rate/threshold: 30
2025-08-21 00:54:07 - INFO - [2999105b-10bc-497e-8931-352c7d9d65e6] Extracted 30 frames successfully. Saving to temporary files...
2025-08-21 00:54:07 - INFO - [2999105b-10bc-497e-8931-352c7d9d65e6] 30 frames saved to temp_videos/2999105b-10bc-497e-8931-352c7d9d65e6
2025-08-21 00:54:07 - INFO - Prompt token length: 2305