chaiml-4d70-fd43-linear-w01

developer_uid: rirv938

submission_id: chaiml-4d70-fd43-linear-w01_v18

model_name: chaiml-4d70-fd43-linear-w01_v18

model_group: ChaiML/4d70-fd43-linear-

status: torndown

timestamp: 2025-12-12T18:26:56+00:00

num_battles: 6450

num_wins: 3318

celo_rating: 1301.63

family_friendly_score: 0.5018

family_friendly_standard_error: 0.007071021991197595

submission_type: basic

model_repo: ChaiML/4d70-fd43-linear-w01

model_architecture: MistralForCausalLM

model_num_parameters: 12772070400.0

best_of: 8

max_input_tokens: 1800

max_output_tokens: 74

reward_model: default

latencies: [{'batch_size': 1, 'throughput': 0.6377690461957166, 'latency_mean': 1.5678654909133911, 'latency_p50': 1.5593345165252686, 'latency_p90': 1.7244777917861938}, {'batch_size': 3, 'throughput': 1.2552412080089532, 'latency_mean': 2.381809192895889, 'latency_p50': 2.3790024518966675, 'latency_p90': 2.642141604423523}, {'batch_size': 5, 'throughput': 1.5614223769590252, 'latency_mean': 3.1917634272575377, 'latency_p50': 3.2110869884490967, 'latency_p90': 3.557779264450073}, {'batch_size': 6, 'throughput': 1.6781224766097025, 'latency_mean': 3.560939282178879, 'latency_p50': 3.544742703437805, 'latency_p90': 4.007159686088562}, {'batch_size': 8, 'throughput': 1.8418243842653528, 'latency_mean': 4.31768835902214, 'latency_p50': 4.333550453186035, 'latency_p90': 4.768724489212036}, {'batch_size': 10, 'throughput': 1.9079875073450945, 'latency_mean': 5.190351094007492, 'latency_p50': 5.1795923709869385, 'latency_p90': 5.882462644577027}]

gpu_counts: {'NVIDIA L40S': 1}

display_name: chaiml-4d70-fd43-linear-w01_v18

ineligible_reason: max_output_tokens!=64

is_internal_developer: True

language_model: ChaiML/4d70-fd43-linear-w01

model_size: 13B

ranking_group: single

throughput_3p7s: 1.72

us_pacific_date: 2025-11-26

win_ratio: 0.5144186046511627

generation_params: {'temperature': 1.05, 'top_p': 1.0, 'min_p': 0.05, 'top_k': 80, 'presence_penalty': 0.08, 'frequency_penalty': 0.08, 'stopping_words': ['\n', '<|im_start|>', '<|im_end|>', '###', '</s>'], 'max_input_tokens': 1800, 'best_of': 8, 'max_output_tokens': 74}

formatter: {'memory_template': '', 'prompt_template': '', 'bot_template': '{bot_name}: {message}\n', 'user_template': '{user_name}: {message}\n', 'response_template': '{bot_name}:', 'truncate_by_message': True}

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-4d70-fd43-linear-w01-v18-mkmlizer
Waiting for job on chaiml-4d70-fd43-linear-w01-v18-mkmlizer to finish
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: ║                                                                     ║
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: ║   ██████   ██████  █████   ████  ████                               ║
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: ║  ░░██████ ██████  ░░███   ███░  ░░███                               ║
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: ║   ░███░█████░███   ░███  ███     ░███                               ║
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: ║   ░███░░███ ░███   ░███████      ░███                               ║
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: ║   ░███ ░░░  ░███   ░███░░███     ░███                               ║
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: ║   ░███      ░███   ░███ ░░███    ░███                               ║
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: ║   █████     █████  █████ ░░████  █████                              ║
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: ║  ░░░░░     ░░░░░  ░░░░░   ░░░░  ░░░░░                               ║
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: ║                                                                     ║
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: ║  Version: 0.30.2                                                    ║
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: ║  Features: FLYWHEEL, CUDA                                           ║
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: ║  Copyright 2023-2025 MK ONE TECHNOLOGIES Inc.                       ║
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: ║  https://mk1.ai                                                     ║
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: ║                                                                     ║
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: ║  The license key for the current software has been verified as      ║
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: ║  belonging to:                                                      ║
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: ║                                                                     ║
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: ║  Chai Research Corp.                                                ║
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: ║  Expiration: 2028-03-31 23:59:59                                    ║
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: ║                                                                     ║
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: Downloaded to shared memory in 22.977s
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: Checking if ChaiML/4d70-fd43-linear-w01 already exists in ChaiML
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpe69xq48s, device:0
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: Saving flywheel model at /dev/shm/model_cache
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: quantized model in 25.364s
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: Processed model ChaiML/4d70-fd43-linear-w01 in 48.342s
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: creating bucket guanaco-mkml-models
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-4d70-fd43-linear-w01-v18/nvidia
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-4d70-fd43-linear-w01-v18/nvidia/tokenizer_config.json
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-4d70-fd43-linear-w01-v18/nvidia/tokenizer.json
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-4d70-fd43-linear-w01-v18/nvidia/flywheel_model.0.safetensors
chaiml-4d70-fd43-linear-w01-v18-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|          | 2/363 [00:05<16:15,  2.70s/it]
Loading 0:   2%|▏         | 6/363 [00:05<04:19,  1.38it/s]
Loading 0:   4%|▍         | 14/363 [00:05<01:24,  4.12it/s]
Loading 0:   6%|▌         | 20/363 [00:05<00:51,  6.72it/s]
Loading 0:   7%|▋         | 26/363 [00:05<00:33, 10.07it/s]
Loading 0:   9%|▉         | 33/363 [00:06<00:22, 14.41it/s]
Loading 0:  11%|█         | 40/363 [00:06<00:17, 18.49it/s]
Loading 0:  13%|█▎        | 46/363 [00:06<00:13, 23.29it/s]
Loading 0:  14%|█▍        | 51/363 [00:06<00:12, 25.63it/s]
Loading 0:  16%|█▋        | 59/363 [00:06<00:08, 34.40it/s]
Loading 0:  18%|█▊        | 65/363 [00:06<00:07, 37.36it/s]
Loading 0:  20%|█▉        | 71/363 [00:06<00:07, 41.40it/s]
Loading 0:  21%|██▏       | 78/363 [00:06<00:06, 42.69it/s]
Loading 0:  24%|██▍       | 87/363 [00:07<00:05, 46.83it/s]
Loading 0:  26%|██▌       | 95/363 [00:07<00:04, 53.87it/s]
Loading 0:  28%|██▊       | 102/363 [00:07<00:04, 54.90it/s]
Loading 0:  30%|██▉       | 108/363 [00:07<00:04, 54.94it/s]
Loading 0:  31%|███▏      | 114/363 [00:07<00:05, 49.42it/s]
Loading 0:  33%|███▎      | 121/363 [00:07<00:05, 43.70it/s]
Loading 0:  35%|███▍      | 127/363 [00:07<00:05, 47.04it/s]
Loading 0:  37%|███▋      | 133/363 [00:08<00:05, 45.82it/s]
Loading 0:  39%|███▉      | 141/363 [00:08<00:04, 47.55it/s]
Loading 0:  41%|████▏     | 150/363 [00:08<00:04, 50.44it/s]
Loading 0:  44%|████▍     | 159/363 [00:08<00:03, 52.82it/s]
Loading 0:  46%|████▋     | 168/363 [00:08<00:03, 53.52it/s]
Loading 0:  49%|████▉     | 177/363 [00:08<00:03, 54.55it/s]
Loading 0:  51%|█████     | 185/363 [00:08<00:02, 60.03it/s]
Loading 0:  53%|█████▎    | 192/363 [00:09<00:02, 59.54it/s]
Loading 0:  55%|█████▍    | 199/363 [00:09<00:02, 59.10it/s]
Loading 0:  57%|█████▋    | 206/363 [00:09<00:03, 40.16it/s]
Loading 0:  59%|█████▊    | 213/363 [00:09<00:03, 41.32it/s]
Loading 0:  61%|██████    | 222/363 [00:09<00:03, 45.29it/s]
Loading 0:  63%|██████▎   | 230/363 [00:09<00:02, 52.11it/s]
Loading 0:  65%|██████▌   | 236/363 [00:10<00:02, 51.14it/s]
Loading 0:  67%|██████▋   | 242/363 [00:10<00:02, 52.60it/s]
Loading 0:  69%|██████▊   | 249/363 [00:10<00:02, 49.63it/s]
Loading 0:  71%|███████   | 257/363 [00:10<00:01, 56.71it/s]
Loading 0:  73%|███████▎  | 264/363 [00:10<00:01, 56.99it/s]
Loading 0:  74%|███████▍  | 270/363 [00:10<00:01, 56.24it/s]
Loading 0:  76%|███████▌  | 276/363 [00:10<00:01, 49.98it/s]
Loading 0:  78%|███████▊  | 283/363 [00:10<00:01, 43.53it/s]
Loading 0:  79%|███████▉  | 288/363 [00:11<00:01, 44.30it/s]
Loading 0:  81%|████████  | 294/363 [00:11<00:01, 42.35it/s]
Loading 0:  83%|████████▎ | 303/363 [00:11<00:01, 46.69it/s]
Loading 0:  86%|████████▌ | 312/363 [00:11<00:01, 49.61it/s]
Loading 0:  88%|████████▊ | 321/363 [00:11<00:00, 51.56it/s]
Loading 0:  91%|█████████ | 330/363 [00:11<00:00, 53.27it/s]
Loading 0:  93%|█████████▎| 338/363 [00:11<00:00, 58.95it/s]
Loading 0:  95%|█████████▌| 345/363 [00:12<00:00, 58.70it/s]
Loading 0:  97%|█████████▋| 352/363 [00:12<00:00, 59.33it/s]
Loading 0:  99%|█████████▉| 359/363 [00:12<00:00, 54.28it/s]
                                                            
Job chaiml-4d70-fd43-linear-w01-v18-mkmlizer completed after 164.74s with status: succeeded
Stopping job with name chaiml-4d70-fd43-linear-w01-v18-mkmlizer
Pipeline stage MKMLizer completed in 165.91s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.28s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service chaiml-4d70-fd43-linear-w01-v18
Waiting for inference service chaiml-4d70-fd43-linear-w01-v18 to be ready
Failed to get response for submission chaiml-prm-ca18v1-pair-d_8033_v1: HTTPConnectionPool(host='chaiml-prm-ca18v1-pair-d-8033-v1-predictor.tenant-chaiml-guanaco.kchai-coreweave-us-east-04a.chaiverse.com', port=80): Read timed out. (read timeout=12.0)
Inference service chaiml-4d70-fd43-linear-w01-v18 ready after 171.6185760498047s
Pipeline stage MKMLDeployer completed in 172.40s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 1.7763950824737549s
Received healthy response to inference request in 1.5849437713623047s
Received healthy response to inference request in 1.9078147411346436s
Received healthy response to inference request in 1.4026308059692383s
Received healthy response to inference request in 1.4545187950134277s
5 requests
0 failed requests
5th percentile: 1.4130084037780761
10th percentile: 1.423386001586914
20th percentile: 1.4441411972045899
30th percentile: 1.4806037902832032
40th percentile: 1.532773780822754
50th percentile: 1.5849437713623047
60th percentile: 1.6615242958068848
70th percentile: 1.7381048202514648
80th percentile: 1.8026790142059326
90th percentile: 1.8552468776702882
95th percentile: 1.8815308094024659
99th percentile: 1.902557954788208
mean time: 1.6252606391906739
Pipeline stage StressChecker completed in 9.60s
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 0.57s
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 0.55s
Shutdown handler de-registered
chaiml-4d70-fd43-linear-w01_v18 status is now deployed due to DeploymentManager action
Shutdown handler registered
run pipeline %s
run pipeline stage %s
Running pipeline stage OfflineFamilyFriendlyScorer
Evaluating %s Family Friendly Score with %s threads
Pipeline stage OfflineFamilyFriendlyScorer completed in 2516.30s
Shutdown handler de-registered
chaiml-4d70-fd43-linear-w01_v18 status is now inactive due to auto deactivation removed underperforming models
chaiml-4d70-fd43-linear-w01_v18 status is now torndown due to DeploymentManager action