chaiml-mistral-24b-2048_64012

developer_uid: rirv938
submission_id: chaiml-mistral-24b-2048_64012_v3
model_name: chaiml-mistral-24b-2048_64012_v3
model_group: ChaiML/mistral_24b_2048_
status: torndown
timestamp: 2026-01-14T16:59:28+00:00
num_battles: 10437
num_wins: 5504
celo_rating: 1314.47
family_friendly_score: 0.48860000000000003
family_friendly_standard_error: 0.007069229661002675
submission_type: basic
model_repo: ChaiML/mistral_24b_2048_opus_v1_5000_40k_merged
model_architecture: MistralForCausalLM
model_num_parameters: 24096691200.0
best_of: 8
max_input_tokens: 2048
max_output_tokens: 96
reward_model: default
latencies: [{'batch_size': 1, 'throughput': 0.34086462436359866, 'latency_mean': 2.933651303052902, 'latency_p50': 2.9391496181488037, 'latency_p90': 3.1503143548965453}, {'batch_size': 2, 'throughput': 0.5274684925938479, 'latency_mean': 3.785536756515503, 'latency_p50': 3.782962203025818, 'latency_p90': 4.004562616348267}, {'batch_size': 3, 'throughput': 0.6592517189542296, 'latency_mean': 4.536740654706955, 'latency_p50': 4.540302395820618, 'latency_p90': 4.895499324798584}, {'batch_size': 4, 'throughput': 0.751677821461648, 'latency_mean': 5.298869621753693, 'latency_p50': 5.307149887084961, 'latency_p90': 5.783296155929565}, {'batch_size': 5, 'throughput': 0.8140369942129445, 'latency_mean': 6.122308971881867, 'latency_p50': 6.095041990280151, 'latency_p90': 6.885562515258789}]
gpu_counts: {'NVIDIA L40S': 1}
display_name: chaiml-mistral-24b-2048_64012_v3
ineligible_reason: max_output_tokens!=64
is_internal_developer: True
language_model: ChaiML/mistral_24b_2048_opus_v1_5000_40k_merged
model_size: 24B
ranking_group: single
throughput_3p7s: 0.51
us_pacific_date: 2026-01-09
win_ratio: 0.5273546038133563
generation_params: {'temperature': 1.0, 'top_p': 0.95, 'min_p': 0.05, 'top_k': 60, 'presence_penalty': 0.1, 'frequency_penalty': 0.0, 'stopping_words': ['<|im_start|>', '<|im_end|>', 'You:', '\n', '</s>', '###'], 'max_input_tokens': 2048, 'best_of': 8, 'max_output_tokens': 96}
formatter: {'memory_template': '', 'prompt_template': '', 'bot_template': '{bot_name}: {message}\n', 'user_template': '{user_name}: {message}\n', 'response_template': '{bot_name}:', 'truncate_by_message': False}
Resubmit model
Shutdown handler not registered because Python interpreter is not running in the main thread
run pipeline %s
run pipeline stage %s
Running pipeline stage MKMLizer
Starting job with name chaiml-mistral-24b-2048-64012-v3-mkmlizer
Waiting for job on chaiml-mistral-24b-2048-64012-v3-mkmlizer to finish
chaiml-mistral-24b-2048-64012-v3-mkmlizer: bash: cannot set terminal process group (-1): Inappropriate ioctl for device
chaiml-mistral-24b-2048-64012-v3-mkmlizer: bash: no job control in this shell
chaiml-mistral-24b-2048-64012-v3-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-mistral-24b-2048-64012-v3-mkmlizer:   __import__('pkg_resources').declare_namespace(__name__)
chaiml-mistral-24b-2048-64012-v3-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-mistral-24b-2048-64012-v3-mkmlizer: ║                                                                     ║
chaiml-mistral-24b-2048-64012-v3-mkmlizer: ║   ██████   ██████  █████   ████  ████                               ║
chaiml-mistral-24b-2048-64012-v3-mkmlizer: ║  ░░██████ ██████  ░░███   ███░  ░░███                               ║
chaiml-mistral-24b-2048-64012-v3-mkmlizer: ║   ░███░█████░███   ░███  ███     ░███                               ║
chaiml-mistral-24b-2048-64012-v3-mkmlizer: ║   ░███░░███ ░███   ░███████      ░███                               ║
chaiml-mistral-24b-2048-64012-v3-mkmlizer: ║   ░███ ░░░  ░███   ░███░░███     ░███                               ║
chaiml-mistral-24b-2048-64012-v3-mkmlizer: ║   ░███      ░███   ░███ ░░███    ░███                               ║
chaiml-mistral-24b-2048-64012-v3-mkmlizer: ║   █████     █████  █████ ░░████  █████                              ║
chaiml-mistral-24b-2048-64012-v3-mkmlizer: ║  ░░░░░     ░░░░░  ░░░░░   ░░░░  ░░░░░                               ║
chaiml-mistral-24b-2048-64012-v3-mkmlizer: ║                                                                     ║
chaiml-mistral-24b-2048-64012-v3-mkmlizer: ║  Version: 0.30.6+torch280                                           ║
chaiml-mistral-24b-2048-64012-v3-mkmlizer: ║  Features: FLYWHEEL, CUDA                                           ║
chaiml-mistral-24b-2048-64012-v3-mkmlizer: ║  Copyright 2023-2025 MK ONE TECHNOLOGIES Inc.                       ║
chaiml-mistral-24b-2048-64012-v3-mkmlizer: ║  https://mk1.ai                                                     ║
chaiml-mistral-24b-2048-64012-v3-mkmlizer: ║                                                                     ║
chaiml-mistral-24b-2048-64012-v3-mkmlizer: ║  The license key for the current software has been verified as      ║
chaiml-mistral-24b-2048-64012-v3-mkmlizer: ║  belonging to:                                                      ║
chaiml-mistral-24b-2048-64012-v3-mkmlizer: ║                                                                     ║
chaiml-mistral-24b-2048-64012-v3-mkmlizer: ║  Chai Research Corp.                                                ║
chaiml-mistral-24b-2048-64012-v3-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
chaiml-mistral-24b-2048-64012-v3-mkmlizer: ║  Expiration: 2028-03-31 23:59:59                                    ║
chaiml-mistral-24b-2048-64012-v3-mkmlizer: ║                                                                     ║
chaiml-mistral-24b-2048-64012-v3-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
chaiml-mistral-24b-2048-64012-v3-mkmlizer: Downloaded to shared memory in 68.989s
chaiml-mistral-24b-2048-64012-v3-mkmlizer: Checking if ChaiML/mistral_24b_2048_opus_v1_5000_40k_merged already exists in ChaiML
chaiml-mistral-24b-2048-64012-v3-mkmlizer: quantizing model to /dev/shm/model_cache, profile:q4, folder:/tmp/tmpu31exazf, device:0
chaiml-mistral-24b-2048-64012-v3-mkmlizer: Saving flywheel model at /dev/shm/model_cache
chaiml-mistral-24b-2048-64012-v3-mkmlizer: 
Loading 0:   0%|          | 0.00/363 [00:00<?, ?it/s]
Loading 0:   1%|          | 3.00/363 [00:01<03:58, 1.51it/s]
Loading 0:   1%|          | 3.00/363 [00:01<03:58, 1.51it/s]
Loading 0:   1%|          | 4.00/363 [00:03<06:20, 1.06s/it]
Loading 0:   1%|          | 4.00/363 [00:03<06:20, 1.06s/it]
Loading 0:   1%|▏         | 5.00/363 [00:06<08:32, 1.43s/it]
Loading 0:   1%|▏         | 5.00/363 [00:06<08:32, 1.43s/it]
Loading 0:   3%|▎         | 11.0/363 [00:07<03:02, 1.93it/s]
Loading 0:   3%|▎         | 11.0/363 [00:07<03:02, 1.93it/s]
Loading 0:   4%|▎         | 13.0/363 [00:09<03:48, 1.53it/s]
Loading 0:   4%|▎         | 13.0/363 [00:09<03:48, 1.53it/s]
Loading 0:   4%|▍         | 14.0/363 [00:11<04:55, 1.18it/s]
Loading 0:   4%|▍         | 14.0/363 [00:11<04:55, 1.18it/s]
Loading 0:   4%|▍         | 15.0/363 [00:13<06:08, 1.06s/it]
Loading 0:   4%|▍         | 15.0/363 [00:13<06:08, 1.06s/it]
Loading 0:   6%|▌         | 21.0/363 [00:15<03:43, 1.53it/s]
Loading 0:   6%|▌         | 21.0/363 [00:15<03:43, 1.53it/s]
Loading 0:   6%|▌         | 22.0/363 [00:17<04:36, 1.23it/s]
Loading 0:   6%|▌         | 22.0/363 [00:17<04:36, 1.23it/s]
Loading 0:   6%|▋         | 23.0/363 [00:19<05:38, 1.00it/s]
Loading 0:   6%|▋         | 23.0/363 [00:19<05:38, 1.00it/s]
Loading 0:   9%|▊         | 31.0/363 [00:20<02:32, 2.18it/s]
Loading 0:   9%|▊         | 31.0/363 [00:20<02:32, 2.18it/s]
Loading 0:   9%|▉         | 34.0/363 [00:22<02:53, 1.90it/s]
Loading 0:   9%|▉         | 34.0/363 [00:22<02:53, 1.90it/s]
Loading 0:  10%|▉         | 35.0/363 [00:24<03:41, 1.48it/s]
Loading 0:  10%|▉         | 35.0/363 [00:24<03:41, 1.48it/s]
Loading 0:  10%|▉         | 36.0/363 [00:26<04:39, 1.17it/s]
Loading 0:  10%|▉         | 36.0/363 [00:26<04:39, 1.17it/s]
Loading 0:  11%|█         | 39.0/363 [00:28<04:10, 1.29it/s]
Loading 0:  11%|█         | 39.0/363 [00:28<04:10, 1.29it/s]
Loading 0:  11%|█         | 40.0/363 [00:30<05:03, 1.07it/s]
Loading 0:  11%|█         | 40.0/363 [00:30<05:03, 1.07it/s]
Loading 0:  11%|█▏        | 41.0/363 [00:32<06:01, 1.12s/it]
Loading 0:  11%|█▏        | 41.0/363 [00:32<06:01, 1.12s/it]
Loading 0:  13%|█▎        | 49.0/363 [00:33<02:27, 2.12it/s]
Loading 0:  13%|█▎        | 49.0/363 [00:33<02:27, 2.12it/s]
Loading 0:  14%|█▍        | 52.0/363 [00:35<02:47, 1.85it/s]
Loading 0:  14%|█▍        | 52.0/363 [00:35<02:47, 1.85it/s]
Loading 0:  15%|█▍        | 53.0/363 [00:37<03:34, 1.45it/s]
Loading 0:  15%|█▍        | 53.0/363 [00:37<03:34, 1.45it/s]
Loading 0:  15%|█▍        | 54.0/363 [00:39<04:29, 1.15it/s]
Loading 0:  15%|█▍        | 54.0/363 [00:39<04:29, 1.15it/s]
Loading 0:  16%|█▌        | 57.0/363 [00:41<04:00, 1.27it/s]
Loading 0:  16%|█▌        | 57.0/363 [00:41<04:00, 1.27it/s]
Loading 0:  16%|█▌        | 58.0/363 [00:43<04:50, 1.05it/s]
Loading 0:  16%|█▌        | 58.0/363 [00:43<04:50, 1.05it/s]
Loading 0:  16%|█▋        | 59.0/363 [00:45<05:47, 1.14s/it]
Loading 0:  16%|█▋        | 59.0/363 [00:45<05:47, 1.14s/it]
Loading 0:  18%|█▊        | 67.0/363 [00:46<02:22, 2.08it/s]
Loading 0:  18%|█▊        | 67.0/363 [00:46<02:22, 2.08it/s]
Loading 0:  19%|█▉        | 70.0/363 [00:49<02:40, 1.83it/s]
Loading 0:  19%|█▉        | 70.0/363 [00:49<02:40, 1.83it/s]
Loading 0:  20%|█▉        | 71.0/363 [00:50<03:23, 1.43it/s]
Loading 0:  20%|█▉        | 71.0/363 [00:50<03:23, 1.43it/s]
Loading 0:  20%|█▉        | 72.0/363 [00:53<04:16, 1.14it/s]
Loading 0:  20%|█▉        | 72.0/363 [00:53<04:16, 1.14it/s]
Loading 0:  21%|██        | 75.0/363 [00:54<03:48, 1.26it/s]
Loading 0:  21%|██        | 75.0/363 [00:54<03:48, 1.26it/s]
Loading 0:  21%|██        | 76.0/363 [00:56<04:35, 1.04it/s]
Loading 0:  21%|██        | 76.0/363 [00:56<04:35, 1.04it/s]
Loading 0:  21%|██        | 77.0/363 [00:58<05:27, 1.14s/it]
Loading 0:  21%|██        | 77.0/363 [00:58<05:27, 1.14s/it]
Loading 0:  23%|██▎       | 85.0/363 [01:00<02:14, 2.07it/s]
Loading 0:  23%|██▎       | 85.0/363 [01:00<02:14, 2.07it/s]
Loading 0:  24%|██▍       | 88.0/363 [01:02<02:31, 1.82it/s]
Loading 0:  24%|██▍       | 88.0/363 [01:02<02:31, 1.82it/s]
Loading 0:  25%|██▍       | 89.0/363 [01:04<03:11, 1.43it/s]
Loading 0:  25%|██▍       | 89.0/363 [01:04<03:11, 1.43it/s]
Loading 0:  25%|██▍       | 90.0/363 [01:06<04:01, 1.13it/s]
Loading 0:  25%|██▍       | 90.0/363 [01:06<04:01, 1.13it/s]
Loading 0:  27%|██▋       | 98.0/363 [01:07<01:55, 2.30it/s]
Loading 0:  27%|██▋       | 98.0/363 [01:07<01:55, 2.30it/s]
Loading 0:  28%|██▊       | 101/363 [01:09<02:10, 2.01it/s] 
Loading 0:  28%|██▊       | 101/363 [01:09<02:10, 2.01it/s]
Loading 0:  28%|██▊       | 102/363 [01:11<02:48, 1.55it/s]
Loading 0:  28%|██▊       | 102/363 [01:11<02:48, 1.55it/s]
Loading 0:  28%|██▊       | 103/363 [01:13<03:34, 1.21it/s]
Loading 0:  28%|██▊       | 103/363 [01:13<03:34, 1.21it/s]
Loading 0:  29%|██▉       | 106/363 [01:15<03:19, 1.29it/s]
Loading 0:  29%|██▉       | 106/363 [01:15<03:19, 1.29it/s]
Loading 0:  29%|██▉       | 107/363 [01:17<04:01, 1.06it/s]
Loading 0:  29%|██▉       | 107/363 [01:17<04:01, 1.06it/s]
Loading 0:  30%|██▉       | 108/363 [01:19<04:47, 1.13s/it]
Loading 0:  30%|██▉       | 108/363 [01:19<04:47, 1.13s/it]
Loading 0:  31%|███       | 111/363 [01:21<03:52, 1.08it/s]
Loading 0:  31%|███       | 111/363 [01:21<03:52, 1.08it/s]
Loading 0:  31%|███       | 112/363 [01:23<04:33, 1.09s/it]
Loading 0:  31%|███       | 112/363 [01:23<04:33, 1.09s/it]
Loading 0:  31%|███       | 113/363 [01:25<05:16, 1.27s/it]
Loading 0:  31%|███       | 113/363 [01:25<05:16, 1.27s/it]
Loading 0:  33%|███▎      | 121/363 [01:26<02:01, 2.00it/s]
Loading 0:  33%|███▎      | 121/363 [01:26<02:01, 2.00it/s]
Loading 0:  34%|███▍      | 124/363 [01:28<02:15, 1.76it/s]
Loading 0:  34%|███▍      | 124/363 [01:28<02:15, 1.76it/s]
Loading 0:  34%|███▍      | 125/363 [01:30<02:51, 1.39it/s]
Loading 0:  34%|███▍      | 125/363 [01:30<02:51, 1.39it/s]
Loading 0:  35%|███▍      | 126/363 [01:32<03:33, 1.11it/s]
Loading 0:  35%|███▍      | 126/363 [01:32<03:33, 1.11it/s]
Loading 0:  36%|███▌      | 129/363 [01:34<03:08, 1.24it/s]
Loading 0:  36%|███▌      | 129/363 [01:34<03:08, 1.24it/s]
Loading 0:  36%|███▌      | 130/363 [01:36<03:46, 1.03it/s]
Loading 0:  36%|███▌      | 130/363 [01:36<03:46, 1.03it/s]
Loading 0:  36%|███▌      | 131/363 [01:38<04:28, 1.16s/it]
Loading 0:  36%|███▌      | 131/363 [01:38<04:28, 1.16s/it]
Loading 0:  38%|███▊      | 139/363 [01:39<01:48, 2.06it/s]
Loading 0:  38%|███▊      | 139/363 [01:39<01:48, 2.06it/s]
Loading 0:  39%|███▉      | 142/363 [01:41<02:01, 1.81it/s]
Loading 0:  39%|███▉      | 142/363 [01:41<02:01, 1.81it/s]
Loading 0:  39%|███▉      | 143/363 [01:43<02:34, 1.42it/s]
Loading 0:  39%|███▉      | 143/363 [01:43<02:34, 1.42it/s]
Loading 0:  40%|███▉      | 144/363 [01:45<03:13, 1.13it/s]
Loading 0:  40%|███▉      | 144/363 [01:45<03:13, 1.13it/s]
Loading 0:  40%|████      | 147/363 [01:47<02:51, 1.26it/s]
Loading 0:  40%|████      | 147/363 [01:47<02:51, 1.26it/s]
Loading 0:  41%|████      | 148/363 [01:49<03:26, 1.04it/s]
Loading 0:  41%|████      | 148/363 [01:49<03:26, 1.04it/s]
Loading 0:  41%|████      | 149/363 [01:51<04:06, 1.15s/it]
Loading 0:  41%|████      | 149/363 [01:51<04:06, 1.15s/it]
Loading 0:  43%|████▎     | 157/363 [01:53<01:39, 2.07it/s]
Loading 0:  43%|████▎     | 157/363 [01:53<01:39, 2.07it/s]
Loading 0:  44%|████▍     | 160/363 [01:55<01:51, 1.81it/s]
Loading 0:  44%|████▍     | 160/363 [01:55<01:51, 1.81it/s]
Loading 0:  44%|████▍     | 161/363 [01:57<02:21, 1.42it/s]
Loading 0:  44%|████▍     | 161/363 [01:57<02:21, 1.42it/s]
Loading 0:  45%|████▍     | 162/363 [01:59<02:57, 1.13it/s]
Loading 0:  45%|████▍     | 162/363 [01:59<02:57, 1.13it/s]
Loading 0:  45%|████▌     | 165/363 [02:01<02:37, 1.26it/s]
Loading 0:  45%|████▌     | 165/363 [02:01<02:37, 1.26it/s]
Loading 0:  46%|████▌     | 166/363 [02:02<03:09, 1.04it/s]
Loading 0:  46%|████▌     | 166/363 [02:02<03:09, 1.04it/s]
Loading 0:  46%|████▌     | 167/363 [02:05<03:45, 1.15s/it]
Loading 0:  46%|████▌     | 167/363 [02:05<03:45, 1.15s/it]
Loading 0:  48%|████▊     | 175/363 [02:06<01:30, 2.07it/s]
Loading 0:  48%|████▊     | 175/363 [02:06<01:30, 2.07it/s]
Loading 0:  49%|████▉     | 178/363 [02:08<01:41, 1.82it/s]
Loading 0:  49%|████▉     | 178/363 [02:08<01:41, 1.82it/s]
Loading 0:  49%|████▉     | 179/363 [02:10<02:09, 1.43it/s]
Loading 0:  49%|████▉     | 179/363 [02:10<02:09, 1.43it/s]
Loading 0:  50%|████▉     | 180/363 [02:12<02:41, 1.13it/s]
Loading 0:  50%|████▉     | 180/363 [02:12<02:41, 1.13it/s]
Loading 0:  50%|█████     | 183/363 [02:14<02:22, 1.26it/s]
Loading 0:  50%|█████     | 183/363 [02:14<02:22, 1.26it/s]
Loading 0:  51%|█████     | 184/363 [02:16<02:51, 1.04it/s]
Loading 0:  51%|█████     | 184/363 [02:16<02:51, 1.04it/s]
Loading 0:  51%|█████     | 185/363 [02:18<03:24, 1.15s/it]
Loading 0:  51%|█████     | 185/363 [02:18<03:24, 1.15s/it]
Loading 0:  53%|█████▎    | 193/363 [02:19<01:22, 2.07it/s]
Loading 0:  53%|█████▎    | 193/363 [02:19<01:22, 2.07it/s]
Loading 0:  54%|█████▍    | 196/363 [02:21<01:31, 1.82it/s]
Loading 0:  54%|█████▍    | 196/363 [02:21<01:31, 1.82it/s]
Loading 0:  54%|█████▍    | 197/363 [02:23<01:56, 1.43it/s]
Loading 0:  54%|█████▍    | 197/363 [02:23<01:56, 1.43it/s]
Loading 0:  55%|█████▍    | 198/363 [02:25<02:25, 1.13it/s]
Loading 0:  55%|█████▍    | 198/363 [02:25<02:25, 1.13it/s]
Loading 0:  55%|█████▌    | 201/363 [02:27<02:08, 1.26it/s]
Loading 0:  55%|█████▌    | 201/363 [02:27<02:08, 1.26it/s]
Loading 0:  56%|█████▌    | 202/363 [02:29<02:34, 1.04it/s]
Loading 0:  56%|█████▌    | 202/363 [02:29<02:34, 1.04it/s]
Loading 0:  56%|█████▌    | 203/363 [02:31<03:03, 1.15s/it]
Loading 0:  56%|█████▌    | 203/363 [02:31<03:03, 1.15s/it]
Loading 0:  58%|█████▊    | 211/363 [02:32<01:13, 2.08it/s]
Loading 0:  58%|█████▊    | 211/363 [02:32<01:13, 2.08it/s]
Loading 0:  59%|█████▉    | 214/363 [02:34<01:21, 1.83it/s]
Loading 0:  59%|█████▉    | 214/363 [02:34<01:21, 1.83it/s]
Loading 0:  59%|█████▉    | 215/363 [02:36<01:43, 1.43it/s]
Loading 0:  59%|█████▉    | 215/363 [02:36<01:43, 1.43it/s]
Loading 0:  60%|█████▉    | 216/363 [02:38<02:09, 1.14it/s]
Loading 0:  60%|█████▉    | 216/363 [02:38<02:09, 1.14it/s]
Loading 0:  60%|██████    | 219/363 [02:40<01:53, 1.27it/s]
Loading 0:  60%|██████    | 219/363 [02:40<01:53, 1.27it/s]
Loading 0:  61%|██████    | 220/363 [02:42<02:16, 1.04it/s]
Loading 0:  61%|██████    | 220/363 [02:42<02:16, 1.04it/s]
Loading 0:  61%|██████    | 221/363 [02:44<02:42, 1.15s/it]
Loading 0:  61%|██████    | 221/363 [02:44<02:42, 1.15s/it]
Loading 0:  63%|██████▎   | 229/363 [02:45<01:04, 2.07it/s]
Loading 0:  63%|██████▎   | 229/363 [02:45<01:04, 2.07it/s]
Loading 0:  64%|██████▍   | 232/363 [02:48<01:11, 1.82it/s]
Loading 0:  64%|██████▍   | 232/363 [02:48<01:11, 1.82it/s]
Loading 0:  64%|██████▍   | 233/363 [02:49<01:30, 1.43it/s]
Loading 0:  64%|██████▍   | 233/363 [02:49<01:30, 1.43it/s]
Loading 0:  64%|██████▍   | 234/363 [02:51<01:53, 1.14it/s]
Loading 0:  64%|██████▍   | 234/363 [02:51<01:53, 1.14it/s]
Loading 0:  65%|██████▌   | 237/363 [02:53<01:39, 1.26it/s]
Loading 0:  65%|██████▌   | 237/363 [02:53<01:39, 1.26it/s]
Loading 0:  66%|██████▌   | 238/363 [02:55<02:00, 1.04it/s]
Loading 0:  66%|██████▌   | 238/363 [02:55<02:00, 1.04it/s]
Loading 0:  66%|██████▌   | 239/363 [02:57<02:22, 1.15s/it]
Loading 0:  66%|██████▌   | 239/363 [02:57<02:22, 1.15s/it]
Loading 0:  68%|██████▊   | 247/363 [02:59<00:55, 2.07it/s]
Loading 0:  68%|██████▊   | 247/363 [02:59<00:55, 2.07it/s]
Loading 0:  69%|██████▉   | 250/363 [03:01<01:02, 1.82it/s]
Loading 0:  69%|██████▉   | 250/363 [03:01<01:02, 1.82it/s]
Loading 0:  69%|██████▉   | 251/363 [03:03<01:18, 1.43it/s]
Loading 0:  69%|██████▉   | 251/363 [03:03<01:18, 1.43it/s]
Loading 0:  69%|██████▉   | 252/363 [03:05<01:37, 1.14it/s]
Loading 0:  69%|██████▉   | 252/363 [03:05<01:37, 1.14it/s]
Loading 0:  70%|███████   | 255/363 [03:07<01:25, 1.26it/s]
Loading 0:  70%|███████   | 255/363 [03:07<01:25, 1.26it/s]
Loading 0:  71%|███████   | 256/363 [03:08<01:42, 1.04it/s]
Loading 0:  71%|███████   | 256/363 [03:08<01:42, 1.04it/s]
Loading 0:  71%|███████   | 257/363 [03:10<02:01, 1.14s/it]
Loading 0:  71%|███████   | 257/363 [03:10<02:01, 1.14s/it]
Loading 0:  73%|███████▎  | 265/363 [03:12<00:47, 2.07it/s]
Loading 0:  73%|███████▎  | 265/363 [03:12<00:47, 2.07it/s]
Loading 0:  74%|███████▍  | 268/363 [03:14<00:52, 1.82it/s]
Loading 0:  74%|███████▍  | 268/363 [03:14<00:52, 1.82it/s]
Loading 0:  74%|███████▍  | 269/363 [03:16<01:05, 1.43it/s]
Loading 0:  74%|███████▍  | 269/363 [03:16<01:05, 1.43it/s]
Loading 0:  74%|███████▍  | 270/363 [03:18<01:22, 1.13it/s]
Loading 0:  74%|███████▍  | 270/363 [03:18<01:22, 1.13it/s]
Loading 0:  75%|███████▌  | 273/363 [03:33<03:37, 2.42s/it]
Loading 0:  75%|███████▌  | 273/363 [03:33<03:37, 2.42s/it]
Loading 0:  75%|███████▌  | 274/363 [03:35<03:28, 2.34s/it]
Loading 0:  75%|███████▌  | 274/363 [03:35<03:28, 2.34s/it]
Loading 0:  76%|███████▌  | 275/363 [03:37<03:21, 2.29s/it]
Loading 0:  76%|███████▌  | 275/363 [03:37<03:21, 2.29s/it]
Loading 0:  78%|███████▊  | 283/363 [03:38<01:08, 1.16it/s]
Loading 0:  78%|███████▊  | 283/363 [03:38<01:08, 1.16it/s]
Loading 0:  79%|███████▉  | 286/363 [03:40<01:03, 1.22it/s]
Loading 0:  79%|███████▉  | 286/363 [03:40<01:03, 1.22it/s]
Loading 0:  79%|███████▉  | 287/363 [03:42<01:10, 1.08it/s]
Loading 0:  79%|███████▉  | 287/363 [03:42<01:10, 1.08it/s]
Loading 0:  79%|███████▉  | 288/363 [03:44<01:20, 1.08s/it]
Loading 0:  79%|███████▉  | 288/363 [03:44<01:20, 1.08s/it]
Loading 0:  80%|████████  | 291/363 [03:46<01:05, 1.09it/s]
Loading 0:  80%|████████  | 291/363 [03:46<01:05, 1.09it/s]
Loading 0:  80%|████████  | 292/363 [03:48<01:15, 1.06s/it]
Loading 0:  80%|████████  | 292/363 [03:48<01:15, 1.06s/it]
Loading 0:  81%|████████  | 293/363 [03:50<01:26, 1.23s/it]
Loading 0:  81%|████████  | 293/363 [03:50<01:26, 1.23s/it]
Loading 0:  83%|████████▎ | 301/363 [03:51<00:31, 1.99it/s]
Loading 0:  83%|████████▎ | 301/363 [03:51<00:31, 1.99it/s]
Loading 0:  84%|████████▎ | 304/363 [03:53<00:32, 1.80it/s]
Loading 0:  84%|████████▎ | 304/363 [03:53<00:32, 1.80it/s]
Loading 0:  84%|████████▍ | 305/363 [03:55<00:41, 1.41it/s]
Loading 0:  84%|████████▍ | 305/363 [03:55<00:41, 1.41it/s]
Loading 0:  84%|████████▍ | 306/363 [03:57<00:50, 1.13it/s]
Loading 0:  84%|████████▍ | 306/363 [03:57<00:50, 1.13it/s]
Loading 0:  85%|████████▌ | 309/363 [03:59<00:43, 1.25it/s]
Loading 0:  85%|████████▌ | 309/363 [03:59<00:43, 1.25it/s]
Loading 0:  85%|████████▌ | 310/363 [04:01<00:51, 1.04it/s]
Loading 0:  85%|████████▌ | 310/363 [04:01<00:51, 1.04it/s]
Loading 0:  86%|████████▌ | 311/363 [04:03<00:59, 1.15s/it]
Loading 0:  86%|████████▌ | 311/363 [04:03<00:59, 1.15s/it]
Loading 0:  88%|████████▊ | 319/363 [04:04<00:20, 2.10it/s]
Loading 0:  88%|████████▊ | 319/363 [04:04<00:20, 2.10it/s]
Loading 0:  89%|████████▊ | 322/363 [04:06<00:22, 1.83it/s]
Loading 0:  89%|████████▊ | 322/363 [04:06<00:22, 1.83it/s]
Loading 0:  89%|████████▉ | 323/363 [04:08<00:27, 1.43it/s]
Loading 0:  89%|████████▉ | 323/363 [04:08<00:27, 1.43it/s]
Loading 0:  89%|████████▉ | 324/363 [04:10<00:34, 1.14it/s]
Loading 0:  89%|████████▉ | 324/363 [04:10<00:34, 1.14it/s]
Loading 0:  90%|█████████ | 327/363 [04:12<00:28, 1.27it/s]
Loading 0:  90%|█████████ | 327/363 [04:12<00:28, 1.27it/s]
Loading 0:  90%|█████████ | 328/363 [04:14<00:33, 1.04it/s]
Loading 0:  90%|█████████ | 328/363 [04:14<00:33, 1.04it/s]
Loading 0:  91%|█████████ | 329/363 [04:16<00:38, 1.14s/it]
Loading 0:  91%|█████████ | 329/363 [04:16<00:38, 1.14s/it]
Loading 0:  93%|█████████▎| 337/363 [04:17<00:12, 2.11it/s]
Loading 0:  93%|█████████▎| 337/363 [04:17<00:12, 2.11it/s]
Loading 0:  94%|█████████▎| 340/363 [04:19<00:12, 1.84it/s]
Loading 0:  94%|█████████▎| 340/363 [04:19<00:12, 1.84it/s]
Loading 0:  94%|█████████▍| 341/363 [04:21<00:15, 1.44it/s]
Loading 0:  94%|█████████▍| 341/363 [04:21<00:15, 1.44it/s]
Loading 0:  94%|█████████▍| 342/363 [04:23<00:18, 1.14it/s]
Loading 0:  94%|█████████▍| 342/363 [04:23<00:18, 1.14it/s]
Loading 0:  95%|█████████▌| 345/363 [04:25<00:14, 1.27it/s]
Loading 0:  95%|█████████▌| 345/363 [04:25<00:14, 1.27it/s]
Loading 0:  95%|█████████▌| 346/363 [04:27<00:16, 1.04it/s]
Loading 0:  95%|█████████▌| 346/363 [04:27<00:16, 1.04it/s]
Loading 0:  96%|█████████▌| 347/363 [04:29<00:18, 1.15s/it]
Loading 0:  96%|█████████▌| 347/363 [04:29<00:18, 1.15s/it]
Loading 0:  98%|█████████▊| 355/363 [04:30<00:03, 2.11it/s]
Loading 0:  98%|█████████▊| 355/363 [04:30<00:03, 2.11it/s]
Loading 0:  99%|█████████▉| 359/363 [04:33<00:02, 1.89it/s]
Loading 0:  99%|█████████▉| 359/363 [04:33<00:02, 1.89it/s]
Loading 0:  99%|█████████▉| 360/363 [04:35<00:02, 1.49it/s]
Loading 0:  99%|█████████▉| 360/363 [04:35<00:02, 1.49it/s]
Loading 0:  99%|█████████▉| 361/363 [04:37<00:01, 1.18it/s]
Loading 0:  99%|█████████▉| 361/363 [04:37<00:01, 1.18it/s]
Loading 0: 100%|██████████| 363/363 [04:37<00:00, 1.18it/s]
                                                           

Loading 0: 100%|██████████| 363/363 [04:37<00:00, 1.31it/s]
chaiml-mistral-24b-2048-64012-v3-mkmlizer: The tokenizer you are loading from '/tmp/tmpu31exazf' with an incorrect regex pattern: https://huggingface.co/mistralai/Mistral-Small-3.1-24B-Instruct-2503/discussions/84#69121093e8b480e709447d5e. This will lead to incorrect tokenization. You should set the `fix_mistral_regex=True` flag when loading this tokenizer to fix this issue.
chaiml-mistral-24b-2048-64012-v3-mkmlizer: quantized model in 283.762s
chaiml-mistral-24b-2048-64012-v3-mkmlizer: Processed model ChaiML/mistral_24b_2048_opus_v1_5000_40k_merged in 352.752s
chaiml-mistral-24b-2048-64012-v3-mkmlizer: creating bucket guanaco-mkml-models
chaiml-mistral-24b-2048-64012-v3-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-mistral-24b-2048-64012-v3-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-mistral-24b-2048-64012-v3/nvidia
chaiml-mistral-24b-2048-64012-v3-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-mistral-24b-2048-64012-v3/nvidia/special_tokens_map.json
chaiml-mistral-24b-2048-64012-v3-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-mistral-24b-2048-64012-v3/nvidia/config.json
chaiml-mistral-24b-2048-64012-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-mistral-24b-2048-64012-v3/nvidia/tokenizer_config.json
chaiml-mistral-24b-2048-64012-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-mistral-24b-2048-64012-v3/nvidia/tokenizer.json
chaiml-mistral-24b-2048-64012-v3-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/chaiml-mistral-24b-2048-64012-v3/nvidia/flywheel_model.1.safetensors
chaiml-mistral-24b-2048-64012-v3-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-mistral-24b-2048-64012-v3/nvidia/flywheel_model.0.safetensors
Job chaiml-mistral-24b-2048-64012-v3-mkmlizer completed after 560.6s with status: succeeded
Stopping job with name chaiml-mistral-24b-2048-64012-v3-mkmlizer
Pipeline stage MKMLizer completed in 561.09s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.14s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service chaiml-mistral-24b-2048-64012-v3
Waiting for inference service chaiml-mistral-24b-2048-64012-v3 to be ready
Inference service chaiml-mistral-24b-2048-64012-v3 ready after 160.66498947143555s
Pipeline stage MKMLDeployer completed in 161.23s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.816030740737915s
Received healthy response to inference request in 2.073552370071411s
Received healthy response to inference request in 2.2385377883911133s
Received healthy response to inference request in 2.409026861190796s
Received healthy response to inference request in 2.212071180343628s
5 requests
0 failed requests
5th percentile: 2.1012561321258545
10th percentile: 2.128959894180298
20th percentile: 2.1843674182891846
30th percentile: 2.217364501953125
40th percentile: 2.2279511451721192
50th percentile: 2.2385377883911133
60th percentile: 2.306733417510986
70th percentile: 2.3749290466308595
80th percentile: 2.4904276371002196
90th percentile: 2.6532291889190676
95th percentile: 2.7346299648284913
99th percentile: 2.7997505855560303
mean time: 2.3498437881469725
Pipeline stage StressChecker completed in 13.15s
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 0.62s
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-mistral-24b-2048_64012_v3 status is now deployed due to DeploymentManager action
Shutdown handler registered
run pipeline %s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Skipping teardown as no inference service was successfully deployed
Pipeline stage MKMLProfilerDeleter completed in 0.09s
run pipeline stage %s
Running pipeline stage MKMLProfilerTemplater
Pipeline stage MKMLProfilerTemplater completed in 0.07s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service chaiml-mistral-24b-2048-64012-v3-profiler
Waiting for inference service chaiml-mistral-24b-2048-64012-v3-profiler to be ready
Inference service chaiml-mistral-24b-2048-64012-v3-profiler ready after 161.30727910995483s
Pipeline stage MKMLProfilerDeployer completed in 161.94s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl --kubeconfig /code/guanaco/guanaco_services/resources/kchai_coreweave_us_east_04a.yaml cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/chaiml-mistral-24b-2b38d18e00e951a6d73e2a42756407033-deplosvqt2:/code/chaiverse_profiler_1767934236 --namespace tenant-chaiml-guanaco
kubectl --kubeconfig /code/guanaco/guanaco_services/resources/kchai_coreweave_us_east_04a.yaml exec -it chaiml-mistral-24b-2b38d18e00e951a6d73e2a42756407033-deplosvqt2 --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1767934236 && python profiles.py profile --best_of_n 8 --auto_batch 5 --batches 1,5,10,15,20,25,30,35,40,45,50,55,60,65,70,75,80,85,90,95,100,105,110,115,120,125,130,135,140,145,150,155,160,165,170,175,180,185,190,195 --samples 200 --input_tokens 2048 --output_tokens 96 --summary /code/chaiverse_profiler_1767934236/summary.json'
%s, retrying in %s seconds...
kubectl --kubeconfig /code/guanaco/guanaco_services/resources/kchai_coreweave_us_east_04a.yaml cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/chaiml-mistral-24b-2b38d18e00e951a6d73e2a42756407033-deplosvqt2:/code/chaiverse_profiler_1767935927 --namespace tenant-chaiml-guanaco
%s, retrying in %s seconds...
kubectl --kubeconfig /code/guanaco/guanaco_services/resources/kchai_coreweave_us_east_04a.yaml cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/chaiml-mistral-24b-2b38d18e00e951a6d73e2a42756407033-deplosvqt2:/code/chaiverse_profiler_1767935928 --namespace tenant-chaiml-guanaco
clean up pipeline due to error=ISVCScriptError('Command failed with error: error: cannot exec into a container in a completed pod; current phase is Failed\n, output: ')
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service chaiml-mistral-24b-2048-64012-v3-profiler is running
Tearing down inference service chaiml-mistral-24b-2048-64012-v3-profiler
Service chaiml-mistral-24b-2048-64012-v3-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 0.58s
Shutdown handler de-registered
Shutdown handler registered
run pipeline %s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service chaiml-mistral-24b-2048-64012-v3-profiler is running
Skipping teardown as no inference service was found
Pipeline stage MKMLProfilerDeleter completed in 0.75s
run pipeline stage %s
Running pipeline stage MKMLProfilerTemplater
Pipeline stage MKMLProfilerTemplater completed in 0.08s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service chaiml-mistral-24b-2048-64012-v3-profiler
Waiting for inference service chaiml-mistral-24b-2048-64012-v3-profiler to be ready
Inference service chaiml-mistral-24b-2048-64012-v3-profiler ready after 161.20915031433105s
Pipeline stage MKMLProfilerDeployer completed in 161.63s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl --kubeconfig /code/guanaco/guanaco_services/resources/kchai_coreweave_us_east_04a.yaml cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/chaiml-mistral-24b-2b38d18e00e951a6d73e2a42756407033-deploh4pnb:/code/chaiverse_profiler_1767936192 --namespace tenant-chaiml-guanaco
kubectl --kubeconfig /code/guanaco/guanaco_services/resources/kchai_coreweave_us_east_04a.yaml exec -it chaiml-mistral-24b-2b38d18e00e951a6d73e2a42756407033-deploh4pnb --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1767936192 && python profiles.py profile --best_of_n 8 --auto_batch 5 --batches 1,5,10,15,20,25,30,35,40,45,50,55,60,65,70,75,80,85,90,95,100,105,110,115,120,125,130,135,140,145,150,155,160,165,170,175,180,185,190,195 --samples 200 --input_tokens 2048 --output_tokens 96 --summary /code/chaiverse_profiler_1767936192/summary.json'
kubectl --kubeconfig /code/guanaco/guanaco_services/resources/kchai_coreweave_us_east_04a.yaml exec -it chaiml-mistral-24b-2b38d18e00e951a6d73e2a42756407033-deploh4pnb --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1767936192/summary.json'
Pipeline stage MKMLProfilerRunner completed in 1787.99s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service chaiml-mistral-24b-2048-64012-v3-profiler is running
Tearing down inference service chaiml-mistral-24b-2048-64012-v3-profiler
Service chaiml-mistral-24b-2048-64012-v3-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 0.68s
Shutdown handler de-registered
chaiml-mistral-24b-2048_64012_v3 status is now inactive due to auto deactivation removed underperforming models
chaiml-mistral-24b-2048_64012_v3 status is now torndown due to DeploymentManager action