Shutdown handler not registered because Python interpreter is not running in the main thread
run pipeline %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
run pipeline stage %s
Running pipeline stage MKMLizer
Starting job with name chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer
Waiting for job on chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer to finish
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: bash: cannot set terminal process group (-1): Inappropriate ioctl for device
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: bash: no job control in this shell
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: /root/miniconda3/envs/nvidia/lib/python3.11/site-packages/mk1/__init__.py:1: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: __import__('pkg_resources').declare_namespace(__name__)
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: ║ ║
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: ║ ██████ ██████ █████ ████ ████ ║
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: ║ ░░██████ ██████ ░░███ ███░ ░░███ ║
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: ║ ░███░█████░███ ░███ ███ ░███ ║
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: ║ ░███░░███ ░███ ░███████ ░███ ║
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: ║ ░███ ░░░ ░███ ░███░░███ ░███ ║
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: ║ ░███ ░███ ░███ ░░███ ░███ ║
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: ║ █████ █████ █████ ░░████ █████ ║
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: ║ ░░░░░ ░░░░░ ░░░░░ ░░░░ ░░░░░ ║
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: ║ ║
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: ║ Version: 0.30.6+torch280 ║
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: ║ Features: FLYWHEEL, CUDA ║
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: ║ Copyright 2023-2025 MK ONE TECHNOLOGIES Inc. ║
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: ║ https://mk1.ai ║
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: ║ ║
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: ║ The license key for the current software has been verified as ║
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: ║ belonging to: ║
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: ║ ║
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: ║ Chai Research Corp. ║
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: ║ Expiration: 2028-03-31 23:59:59 ║
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: ║ ║
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: Downloaded to shared memory in 15.727s
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: Checking if ChaiML/prm_kimi_v0_180k_default8b-cosine-lr1e6g16 already exists in ChaiML
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: quantizing model to /dev/shm/model_cache, profile:t0, folder:/tmp/tmpm_8ij07i, device:0
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer:
Loading 0: 0%| | 0.00/291 [00:00<?, ?it/s]
Loading 0: 16%|█▌ | 46.0/291 [00:01<00:05, 43.3it/s]
Loading 0: 16%|█▌ | 46.0/291 [00:01<00:05, 43.3it/s]
Loading 0: 29%|██▉ | 85.0/291 [00:02<00:05, 40.5it/s]
Loading 0: 29%|██▉ | 85.0/291 [00:02<00:05, 40.5it/s]
Loading 0: 44%|████▎ | 127/291 [00:03<00:04, 40.0it/s]
Loading 0: 44%|████▎ | 127/291 [00:03<00:04, 40.0it/s]
Loading 0: 59%|█████▉ | 172/291 [00:04<00:02, 41.7it/s]
Loading 0: 59%|█████▉ | 172/291 [00:04<00:02, 41.7it/s]
Loading 0: 73%|███████▎ | 211/291 [00:05<00:01, 40.6it/s]
Loading 0: 73%|███████▎ | 211/291 [00:05<00:01, 40.6it/s]
Loading 0: 88%|████████▊ | 257/291 [00:06<00:00, 41.3it/s]
Loading 0: 88%|████████▊ | 257/291 [00:06<00:00, 41.3it/s]
Loading 0: 100%|██████████| 291/291 [00:07<00:00, 39.9it/s]
Loading 0: 100%|██████████| 291/291 [00:07<00:00, 39.9it/s]
Loading 0: 100%|██████████| 291/291 [00:07<00:00, 40.6it/s]
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: quantized model in 18.232s
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: Processed model ChaiML/prm_kimi_v0_180k_default8b-cosine-lr1e6g16 in 33.960s
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: creating bucket guanaco-mkml-models
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-prm-kimi-v0-180k-25084-v2/nvidia
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-prm-kimi-v0-180k-25084-v2/nvidia/config.json
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-prm-kimi-v0-180k-25084-v2/nvidia/special_tokens_map.json
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-prm-kimi-v0-180k-25084-v2/nvidia/tokenizer_config.json
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-prm-kimi-v0-180k-25084-v2/nvidia/tokenizer.json
Unable to record family friendly update due to error: Invalid JSON input: Expecting value: line 1 column 1 (char 0)
chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-prm-kimi-v0-180k-25084-v2/nvidia/flywheel_model.0.safetensors
Job chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer completed after 93.62s with status: succeeded
Stopping job with name chaiml-prm-kimi-v0-180k-25084-v2-mkmlizer
Pipeline stage MKMLizer completed in 94.59s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.22s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service chaiml-prm-kimi-v0-180k-25084-v2
Waiting for inference service chaiml-prm-kimi-v0-180k-25084-v2 to be ready
Inference service chaiml-prm-kimi-v0-180k-25084-v2 ready after 150.83472394943237s
Pipeline stage MKMLDeployer completed in 151.38s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 4.5242884159088135s
Received healthy response to inference request in 2.874242067337036s
Received healthy response to inference request in 4.984886884689331s
Received healthy response to inference request in 4.761233806610107s
Received healthy response to inference request in 4.557452201843262s
5 requests
0 failed requests
5th percentile: 3.2042513370513914
10th percentile: 3.534260606765747
20th percentile: 4.194279146194458
30th percentile: 4.530921173095703
40th percentile: 4.544186687469482
50th percentile: 4.557452201843262
60th percentile: 4.63896484375
70th percentile: 4.720477485656739
80th percentile: 4.805964422225952
90th percentile: 4.895425653457641
95th percentile: 4.940156269073486
99th percentile: 4.975940761566162
mean time: 4.34042067527771
Pipeline stage StressChecker completed in 23.15s
run pipeline stage %s
Running pipeline stage OfflineFamilyFriendlyTriggerPipeline
run_pipeline:run_in_cloud %s
starting trigger_guanaco_pipeline args=%s
triggered trigger_guanaco_pipeline args=%s
Pipeline stage OfflineFamilyFriendlyTriggerPipeline completed in 0.61s
run pipeline stage %s
Running pipeline stage TriggerMKMLProfilingPipeline
run_pipeline:run_in_cloud %s
starting trigger_guanaco_pipeline args=%s
triggered trigger_guanaco_pipeline args=%s
Pipeline stage TriggerMKMLProfilingPipeline completed in 0.61s
Shutdown handler de-registered
chaiml-prm-kimi-v0-180k_25084_v2 status is now deployed due to DeploymentManager action
chaiml-prm-kimi-v0-180k_25084_v2 status is now inactive due to auto deactivation removed underperforming models