chaiml-ca18-v1-4-dpo-cosine

developer_uid: chai_backend_admin

submission_id: chaiml-ca18-v1-4-dpo-cosine_v1

model_name: training123

model_group: ChaiML/ca18-v1-4-dpo_cos

status: torndown

timestamp: 2025-11-19T09:01:34+00:00

num_battles: 5686

num_wins: 2989

celo_rating: 1305.17

family_friendly_score: 0.506

family_friendly_standard_error: 0.007070558676653493

submission_type: basic

model_repo: ChaiML/ca18-v1-4-dpo_cosine

model_architecture: MistralForCausalLM

model_num_parameters: 12772070400.0

best_of: 8

max_input_tokens: 2048

max_output_tokens: 64

reward_model: default

latencies: [{'batch_size': 1, 'throughput': 0.6860572242233619, 'latency_mean': 1.4574991607666015, 'latency_p50': 1.4633969068527222, 'latency_p90': 1.6008193969726563}, {'batch_size': 3, 'throughput': 1.3065257237538728, 'latency_mean': 2.2873734855651855, 'latency_p50': 2.271301746368408, 'latency_p90': 2.5974424839019776}, {'batch_size': 5, 'throughput': 1.6026960667146601, 'latency_mean': 3.105960165262222, 'latency_p50': 3.0769137144088745, 'latency_p90': 3.5206470251083375}, {'batch_size': 6, 'throughput': 1.6989301641230876, 'latency_mean': 3.519041428565979, 'latency_p50': 3.5020397901535034, 'latency_p90': 3.981267213821411}, {'batch_size': 8, 'throughput': 1.8407646324388491, 'latency_mean': 4.3204670488834385, 'latency_p50': 4.328105211257935, 'latency_p90': 4.944224858283996}, {'batch_size': 10, 'throughput': 1.9128662197196047, 'latency_mean': 5.199235743284225, 'latency_p50': 5.189219355583191, 'latency_p90': 5.94719545841217}]

gpu_counts: {'NVIDIA L40S': 1}

display_name: training123

is_internal_developer: True

language_model: ChaiML/ca18-v1-4-dpo_cosine

model_size: 13B

ranking_group: single

throughput_3p7s: 1.74

us_pacific_date: 2025-11-16

win_ratio: 0.5256771016531833

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': ['\n', 'You:', 'User:', '</s>'], '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
run pipeline %s
run pipeline stage %s
Running pipeline stage MKMLizer
Starting job with name chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer
Waiting for job on chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer to finish
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: ║                                                                     ║
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: ║   ██████   ██████  █████   ████  ████                               ║
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: ║  ░░██████ ██████  ░░███   ███░  ░░███                               ║
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: ║   ░███░█████░███   ░███  ███     ░███                               ║
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: ║   ░███░░███ ░███   ░███████      ░███                               ║
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: ║   ░███ ░░░  ░███   ░███░░███     ░███                               ║
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: ║   ░███      ░███   ░███ ░░███    ░███                               ║
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: ║   █████     █████  █████ ░░████  █████                              ║
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: ║  ░░░░░     ░░░░░  ░░░░░   ░░░░  ░░░░░                               ║
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: ║                                                                     ║
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: ║  Version: 0.30.2                                                    ║
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: ║  Features: FLYWHEEL, CUDA                                           ║
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: ║  Copyright 2023-2025 MK ONE TECHNOLOGIES Inc.                       ║
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: ║  https://mk1.ai                                                     ║
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: ║                                                                     ║
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: ║  belonging to:                                                      ║
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: ║                                                                     ║
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: ║  Chai Research Corp.                                                ║
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: ║  Expiration: 2028-03-31 23:59:59                                    ║
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: ║                                                                     ║
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: Downloaded to shared memory in 52.998s
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: Checking if ChaiML/ca18-v1-4-dpo_cosine already exists in ChaiML
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpw4f3j34y, device:0
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: quantized model in 25.826s
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: Processed model ChaiML/ca18-v1-4-dpo_cosine in 78.824s
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: creating bucket guanaco-mkml-models
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-ca18-v1-4-dpo-cosine-v1/nvidia/config.json
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-ca18-v1-4-dpo-cosine-v1/nvidia/special_tokens_map.json
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-ca18-v1-4-dpo-cosine-v1/nvidia/tokenizer_config.json
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-ca18-v1-4-dpo-cosine-v1/nvidia/tokenizer.json
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-ca18-v1-4-dpo-cosine-v1/nvidia/flywheel_model.0.safetensors
chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|▏         | 5/363 [00:00<00:10, 35.23it/s]
Loading 0:   4%|▍         | 14/363 [00:00<00:07, 48.35it/s]
Loading 0:   6%|▋         | 23/363 [00:00<00:06, 53.24it/s]
Loading 0:   9%|▉         | 32/363 [00:00<00:06, 54.83it/s]
Loading 0:  11%|█▏        | 41/363 [00:00<00:05, 55.29it/s]
Loading 0:  14%|█▍        | 50/363 [00:00<00:05, 56.38it/s]
Loading 0:  17%|█▋        | 60/363 [00:01<00:04, 61.77it/s]
Loading 0:  18%|█▊        | 67/363 [00:01<00:06, 47.15it/s]
Loading 0:  20%|██        | 73/363 [00:01<00:06, 45.11it/s]
Loading 0:  23%|██▎       | 82/363 [00:01<00:05, 47.55it/s]
Loading 0:  25%|██▍       | 90/363 [00:01<00:05, 53.90it/s]
Loading 0:  26%|██▋       | 96/363 [00:01<00:05, 51.95it/s]
Loading 0:  28%|██▊       | 102/363 [00:01<00:04, 53.16it/s]
Loading 0:  30%|███       | 110/363 [00:02<00:04, 54.70it/s]
Loading 0:  32%|███▏      | 116/363 [00:02<00:04, 52.24it/s]
Loading 0:  34%|███▎      | 122/363 [00:02<00:04, 53.97it/s]
Loading 0:  35%|███▌      | 128/363 [00:02<00:04, 49.33it/s]
Loading 0:  37%|███▋      | 135/363 [00:02<00:04, 54.31it/s]
Loading 0:  39%|███▉      | 141/363 [00:02<00:04, 51.99it/s]
Loading 0:  40%|████      | 147/363 [00:02<00:05, 39.73it/s]
Loading 0:  42%|████▏     | 152/363 [00:03<00:05, 41.39it/s]
Loading 0:  44%|████▎     | 158/363 [00:03<00:05, 40.21it/s]
Loading 0:  46%|████▌     | 166/363 [00:03<00:04, 49.03it/s]
Loading 0:  47%|████▋     | 172/363 [00:03<00:04, 44.10it/s]
Loading 0:  49%|████▉     | 177/363 [00:03<00:04, 39.94it/s]
Loading 0:  51%|█████     | 184/363 [00:03<00:03, 46.61it/s]
Loading 0:  52%|█████▏    | 190/363 [00:03<00:03, 46.78it/s]
Loading 0:  54%|█████▎    | 195/363 [00:03<00:03, 47.06it/s]
Loading 0:  56%|█████▌    | 203/363 [00:04<00:03, 48.18it/s]
Loading 0:  58%|█████▊    | 211/363 [00:04<00:02, 55.63it/s]
Loading 0:  60%|█████▉    | 217/363 [00:04<00:02, 52.60it/s]
Loading 0:  61%|██████▏   | 223/363 [00:04<00:03, 43.09it/s]
Loading 0:  63%|██████▎   | 228/363 [00:04<00:03, 43.14it/s]
Loading 0:  64%|██████▍   | 233/363 [00:04<00:02, 44.09it/s]
Loading 0:  66%|██████▌   | 239/363 [00:04<00:02, 42.01it/s]
Loading 0:  68%|██████▊   | 247/363 [00:05<00:02, 50.43it/s]
Loading 0:  70%|██████▉   | 253/363 [00:05<00:02, 49.08it/s]
Loading 0:  71%|███████▏  | 259/363 [00:05<00:02, 49.66it/s]
Loading 0:  73%|███████▎  | 265/363 [00:05<00:01, 52.28it/s]
Loading 0:  75%|███████▍  | 271/363 [00:05<00:01, 50.77it/s]
Loading 0:  76%|███████▋  | 277/363 [00:05<00:01, 52.45it/s]
Loading 0:  78%|███████▊  | 284/363 [00:05<00:01, 49.29it/s]
Loading 0:  80%|████████  | 292/363 [00:05<00:01, 56.49it/s]
Loading 0:  82%|████████▏ | 298/363 [00:06<00:01, 54.13it/s]
Loading 0:  84%|████████▎ | 304/363 [00:06<00:01, 32.21it/s]
Loading 0:  85%|████████▌ | 309/363 [00:06<00:01, 35.34it/s]
Loading 0:  87%|████████▋ | 314/363 [00:06<00:01, 36.45it/s]
Loading 0:  88%|████████▊ | 321/363 [00:06<00:01, 38.96it/s]
Loading 0:  91%|█████████ | 330/363 [00:06<00:00, 44.83it/s]
Loading 0:  93%|█████████▎| 339/363 [00:07<00:00, 48.69it/s]
Loading 0:  96%|█████████▌| 348/363 [00:07<00:00, 51.27it/s]
Loading 0:  98%|█████████▊| 357/363 [00:07<00:00, 53.01it/s]
                                                            
Job chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer completed after 113.44s with status: succeeded
Stopping job with name chaiml-ca18-v1-4-dpo-cosine-v1-mkmlizer
Pipeline stage MKMLizer completed in 114.24s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.38s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service chaiml-ca18-v1-4-dpo-cosine-v1
Waiting for inference service chaiml-ca18-v1-4-dpo-cosine-v1 to be ready
Inference service chaiml-ca18-v1-4-dpo-cosine-v1 ready after 170.91201663017273s
Pipeline stage MKMLDeployer completed in 171.43s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 1.6789488792419434s
Received healthy response to inference request in 1.2929890155792236s
Received healthy response to inference request in 1.6799335479736328s
Received healthy response to inference request in 1.2385191917419434s
Received healthy response to inference request in 1.4336268901824951s
5 requests
0 failed requests
5th percentile: 1.2494131565093993
10th percentile: 1.2603071212768555
20th percentile: 1.2820950508117677
30th percentile: 1.3211165904998778
40th percentile: 1.3773717403411865
50th percentile: 1.4336268901824951
60th percentile: 1.5317556858062744
70th percentile: 1.6298844814300537
80th percentile: 1.6791458129882812
90th percentile: 1.679539680480957
95th percentile: 1.679736614227295
99th percentile: 1.6798941612243652
mean time: 1.4648035049438477
Pipeline stage StressChecker completed in 8.82s
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 0.67s
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 0.83s
Shutdown handler de-registered
chaiml-ca18-v1-4-dpo-cosine_v1 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 0.08s
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-ca18-v1-4-dpo-cosine-v1-profiler
Waiting for inference service chaiml-ca18-v1-4-dpo-cosine-v1-profiler to be ready
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 2480.74s
Shutdown handler de-registered
chaiml-ca18-v1-4-dpo-cosine_v1 status is now inactive due to auto deactivation removed underperforming models
chaiml-ca18-v1-4-dpo-cosine_v1 status is now torndown due to DeploymentManager action