chaiml-ir-v4-1-dpo-ir31_39115

developer_uid: chai_backend_admin

submission_id: chaiml-ir-v4-1-dpo-ir31_39115_v1

model_name: training123

model_group: ChaiML/ir-v4-1-dpo_ir31-

status: torndown

timestamp: 2025-11-07T11:21:19+00:00

num_battles: 6288

num_wins: 3352

celo_rating: 1303.03

family_friendly_score: 0.4256

family_friendly_standard_error: 0.006992347817435857

submission_type: basic

model_repo: ChaiML/ir-v4-1-dpo_ir31-lr5e6b01

model_architecture: MistralForCausalLM

model_num_parameters: 24096691200.0

best_of: 8

max_input_tokens: 2048

max_output_tokens: 64

reward_model: default

latencies: [{'batch_size': 1, 'throughput': 0.20746751996941298, 'latency_mean': 4.819936684370041, 'latency_p50': 4.571645259857178, 'latency_p90': 6.149334883689879}, {'batch_size': 2, 'throughput': 0.5698756120319718, 'latency_mean': 3.5062252533435823, 'latency_p50': 3.520747423171997, 'latency_p90': 3.7948803901672363}, {'batch_size': 3, 'throughput': 0.7034788355471171, 'latency_mean': 4.25288267493248, 'latency_p50': 4.27854311466217, 'latency_p90': 4.616076588630676}, {'batch_size': 4, 'throughput': 0.7937891714322864, 'latency_mean': 5.017153769731522, 'latency_p50': 5.042113184928894, 'latency_p90': 5.60896291732788}, {'batch_size': 5, 'throughput': 0.8555108139126576, 'latency_mean': 5.827774149179459, 'latency_p50': 5.849305152893066, 'latency_p90': 6.692305636405944}]

gpu_counts: {'NVIDIA L40S': 1}

display_name: training123

ineligible_reason: num_battles<10000

is_internal_developer: True

language_model: ChaiML/ir-v4-1-dpo_ir31-lr5e6b01

model_size: 24B

ranking_group: single

throughput_3p7s: 0.27

us_pacific_date: 2025-11-04

win_ratio: 0.5330788804071247

generation_params: {'temperature': 1.0, 'top_p': 1.0, 'min_p': 0.0, 'top_k': 80, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['</s>', 'You:', '\n', 'User:'], 'max_input_tokens': 2048, 'best_of': 8, 'max_output_tokens': 64}

formatter: {'memory_template': "{bot_name}'s Persona: {memory}\n####\n", 'prompt_template': '{prompt}\n<START>\n', '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
Connection pool is full, discarding connection: %s. Connection pool size: %s
run pipeline %s
run pipeline stage %s
Running pipeline stage MKMLizer
Starting job with name chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer
Waiting for job on chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer to finish
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: ║                                                                     ║
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: ║   ██████   ██████  █████   ████  ████                               ║
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: ║  ░░██████ ██████  ░░███   ███░  ░░███                               ║
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: ║   ░███░█████░███   ░███  ███     ░███                               ║
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: ║   ░███░░███ ░███   ░███████      ░███                               ║
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: ║   ░███ ░░░  ░███   ░███░░███     ░███                               ║
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: ║   ░███      ░███   ░███ ░░███    ░███                               ║
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: ║   █████     █████  █████ ░░████  █████                              ║
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: ║  ░░░░░     ░░░░░  ░░░░░   ░░░░  ░░░░░                               ║
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: ║                                                                     ║
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: ║  Version: 0.30.2                                                    ║
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: ║  Features: FLYWHEEL, CUDA                                           ║
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: ║  Copyright 2023-2025 MK ONE TECHNOLOGIES Inc.                       ║
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: ║  https://mk1.ai                                                     ║
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: ║                                                                     ║
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: ║  belonging to:                                                      ║
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: ║                                                                     ║
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: ║  Chai Research Corp.                                                ║
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: ║  Expiration: 2028-03-31 23:59:59                                    ║
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: ║                                                                     ║
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: Downloaded to shared memory in 60.480s
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: Checking if ChaiML/ir-v4-1-dpo_ir31-lr5e6b01 already exists in ChaiML
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpbt7o70j_, device:0
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: quantized model in 42.621s
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: Processed model ChaiML/ir-v4-1-dpo_ir31-lr5e6b01 in 103.102s
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: creating bucket guanaco-mkml-models
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-ir-v4-1-dpo-ir31-39115-v1/nvidia
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-ir-v4-1-dpo-ir31-39115-v1/nvidia/config.json
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-ir-v4-1-dpo-ir31-39115-v1/nvidia/special_tokens_map.json
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-ir-v4-1-dpo-ir31-39115-v1/nvidia/tokenizer_config.json
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-ir-v4-1-dpo-ir31-39115-v1/nvidia/tokenizer.json
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/chaiml-ir-v4-1-dpo-ir31-39115-v1/nvidia/flywheel_model.1.safetensors
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-ir-v4-1-dpo-ir31-39115-v1/nvidia/flywheel_model.0.safetensors
chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|          | 4/363 [00:00<00:09, 36.70it/s]
Loading 0:   2%|▏         | 8/363 [00:00<00:11, 30.73it/s]
Loading 0:   4%|▎         | 13/363 [00:00<00:10, 32.56it/s]
Loading 0:   5%|▍         | 17/363 [00:00<00:10, 33.67it/s]
Loading 0:   6%|▌         | 22/363 [00:00<00:09, 34.21it/s]
Loading 0:   7%|▋         | 26/363 [00:00<00:09, 34.50it/s]
Loading 0:   9%|▉         | 32/363 [00:00<00:08, 40.28it/s]
Loading 0:  10%|█         | 37/363 [00:01<00:13, 24.64it/s]
Loading 0:  11%|█▏        | 41/363 [00:01<00:13, 23.19it/s]
Loading 0:  13%|█▎        | 48/363 [00:01<00:10, 30.85it/s]
Loading 0:  14%|█▍        | 52/363 [00:01<00:10, 30.04it/s]
Loading 0:  16%|█▌        | 57/363 [00:01<00:09, 33.65it/s]
Loading 0:  17%|█▋        | 61/363 [00:01<00:09, 30.74it/s]
Loading 0:  18%|█▊        | 65/363 [00:02<00:09, 31.99it/s]
Loading 0:  19%|█▉        | 70/363 [00:02<00:10, 28.03it/s]
Loading 0:  20%|██        | 74/363 [00:02<00:11, 25.45it/s]
Loading 0:  22%|██▏       | 79/363 [00:02<00:09, 29.40it/s]
Loading 0:  23%|██▎       | 83/363 [00:02<00:09, 30.70it/s]
Loading 0:  24%|██▍       | 87/363 [00:02<00:10, 27.15it/s]
Loading 0:  25%|██▌       | 92/363 [00:03<00:10, 26.50it/s]
Loading 0:  27%|██▋       | 99/363 [00:03<00:07, 33.94it/s]
Loading 0:  28%|██▊       | 103/363 [00:03<00:08, 32.02it/s]
Loading 0:  29%|██▉       | 107/363 [00:03<00:09, 27.23it/s]
Loading 0:  31%|███       | 112/363 [00:03<00:08, 29.83it/s]
Loading 0:  32%|███▏      | 116/363 [00:03<00:08, 30.63it/s]
Loading 0:  33%|███▎      | 120/363 [00:03<00:07, 32.54it/s]
Loading 0:  34%|███▍      | 124/363 [00:04<00:07, 31.27it/s]
Loading 0:  36%|███▌      | 129/363 [00:04<00:06, 34.75it/s]
Loading 0:  37%|███▋      | 133/363 [00:04<00:07, 32.43it/s]
Loading 0:  38%|███▊      | 138/363 [00:04<00:06, 35.64it/s]
Loading 0:  39%|███▉      | 142/363 [00:04<00:06, 32.96it/s]
Loading 0:  41%|████      | 149/363 [00:04<00:05, 39.37it/s]
Loading 0:  42%|████▏     | 154/363 [00:05<00:08, 23.65it/s]
Loading 0:  44%|████▎     | 158/363 [00:05<00:09, 22.30it/s]
Loading 0:  45%|████▌     | 165/363 [00:05<00:06, 29.11it/s]
Loading 0:  47%|████▋     | 169/363 [00:05<00:06, 28.70it/s]
Loading 0:  48%|████▊     | 174/363 [00:05<00:05, 31.84it/s]
Loading 0:  49%|████▉     | 178/363 [00:05<00:06, 30.63it/s]
Loading 0:  50%|█████     | 182/363 [00:06<00:05, 31.69it/s]
Loading 0:  52%|█████▏    | 187/363 [00:06<00:06, 28.96it/s]
Loading 0:  53%|█████▎    | 191/363 [00:06<00:06, 28.41it/s]
Loading 0:  54%|█████▎    | 195/363 [00:06<00:06, 25.83it/s]
Loading 0:  55%|█████▌    | 201/363 [00:19<02:14,  1.20it/s]
Loading 0:  56%|█████▌    | 203/363 [00:19<01:55,  1.38it/s]
Loading 0:  57%|█████▋    | 208/363 [00:19<01:13,  2.09it/s]
Loading 0:  58%|█████▊    | 211/363 [00:20<00:57,  2.64it/s]
Loading 0:  59%|█████▉    | 214/363 [00:20<00:43,  3.41it/s]
Loading 0:  60%|██████    | 218/363 [00:20<00:30,  4.81it/s]
Loading 0:  61%|██████    | 222/363 [00:20<00:21,  6.61it/s]
Loading 0:  62%|██████▏   | 226/363 [00:20<00:16,  8.19it/s]
Loading 0:  63%|██████▎   | 230/363 [00:20<00:13, 10.06it/s]
Loading 0:  65%|██████▌   | 237/363 [00:20<00:08, 15.70it/s]
Loading 0:  66%|██████▋   | 241/363 [00:21<00:06, 17.76it/s]
Loading 0:  68%|██████▊   | 246/363 [00:21<00:05, 21.91it/s]
Loading 0:  69%|██████▉   | 250/363 [00:21<00:04, 23.07it/s]
Loading 0:  70%|███████   | 255/363 [00:21<00:03, 27.34it/s]
Loading 0:  71%|███████▏  | 259/363 [00:21<00:03, 27.60it/s]
Loading 0:  73%|███████▎  | 266/363 [00:21<00:02, 34.57it/s]
Loading 0:  75%|███████▍  | 271/363 [00:21<00:03, 25.29it/s]
Loading 0:  76%|███████▌  | 275/363 [00:22<00:03, 23.69it/s]
Loading 0:  78%|███████▊  | 282/363 [00:22<00:02, 30.04it/s]
Loading 0:  79%|███████▉  | 286/363 [00:22<00:02, 29.11it/s]
Loading 0:  80%|████████  | 291/363 [00:22<00:02, 32.41it/s]
Loading 0:  81%|████████▏ | 295/363 [00:22<00:02, 30.77it/s]
Loading 0:  82%|████████▏ | 299/363 [00:22<00:02, 31.99it/s]
Loading 0:  84%|████████▎ | 304/363 [00:23<00:01, 29.51it/s]
Loading 0:  85%|████████▍ | 308/363 [00:23<00:01, 28.80it/s]
Loading 0:  86%|████████▌ | 312/363 [00:23<00:01, 26.22it/s]
Loading 0:  88%|████████▊ | 318/363 [00:23<00:01, 31.99it/s]
Loading 0:  89%|████████▊ | 322/363 [00:23<00:01, 29.17it/s]
Loading 0:  90%|█████████ | 327/363 [00:23<00:01, 32.39it/s]
Loading 0:  91%|█████████ | 331/363 [00:23<00:01, 31.37it/s]
Loading 0:  92%|█████████▏| 335/363 [00:24<00:00, 32.60it/s]
Loading 0:  93%|█████████▎| 339/363 [00:24<00:00, 33.31it/s]
Loading 0:  94%|█████████▍| 343/363 [00:24<00:00, 20.32it/s]
Loading 0:  96%|█████████▌| 348/363 [00:24<00:00, 21.81it/s]
Loading 0:  98%|█████████▊| 355/363 [00:24<00:00, 29.17it/s]
Loading 0:  99%|█████████▉| 359/363 [00:25<00:00, 28.81it/s]
                                                            
Job chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer completed after 174.98s with status: succeeded
Stopping job with name chaiml-ir-v4-1-dpo-ir31-39115-v1-mkmlizer
Pipeline stage MKMLizer completed in 176.02s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.18s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service chaiml-ir-v4-1-dpo-ir31-39115-v1
Waiting for inference service chaiml-ir-v4-1-dpo-ir31-39115-v1 to be ready
Unable to record family friendly update due to error: Invalid JSON input: Expecting value: line 1 column 1 (char 0)
Inference service chaiml-ir-v4-1-dpo-ir31-39115-v1 ready after 140.61709356307983s
Pipeline stage MKMLDeployer completed in 141.29s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.928572654724121s
Received healthy response to inference request in 2.2829864025115967s
Received healthy response to inference request in 2.377171754837036s
Received healthy response to inference request in 2.119889974594116s
Received healthy response to inference request in 2.1449127197265625s
5 requests
0 failed requests
5th percentile: 2.1248945236206054
10th percentile: 2.1298990726470945
20th percentile: 2.1399081707000733
30th percentile: 2.1725274562835692
40th percentile: 2.227756929397583
50th percentile: 2.2829864025115967
60th percentile: 2.3206605434417726
70th percentile: 2.358334684371948
80th percentile: 2.487451934814453
90th percentile: 2.7080122947692873
95th percentile: 2.818292474746704
99th percentile: 2.9065166187286375
mean time: 2.3707067012786864
Pipeline stage StressChecker completed in 13.68s
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 0.76s
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 0.71s
Shutdown handler de-registered
chaiml-ir-v4-1-dpo-ir31_39115_v1 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 3078.98s
Shutdown handler de-registered
chaiml-ir-v4-1-dpo-ir31_39115_v1 status is now inactive due to auto deactivation removed underperforming models
chaiml-ir-v4-1-dpo-ir31_39115_v1 status is now torndown due to DeploymentManager action