chaiml-ca18-v1-dpo-cosine

developer_uid: chai_backend_admin

submission_id: chaiml-ca18-v1-dpo-cosine_v1

model_name: training123

model_group: ChaiML/ca18-v1-dpo_cosin

status: torndown

timestamp: 2025-11-18T10:21:27+00:00

num_battles: 7716

num_wins: 3997

celo_rating: 1297.95

family_friendly_score: 0.5152

family_friendly_standard_error: 0.007067799657602075

submission_type: basic

model_repo: ChaiML/ca18-v1-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': 0.6978387487207965, 'latency_mean': 1.4328984344005584, 'latency_p50': 1.421125888824463, 'latency_p90': 1.6019403696060182}, {'batch_size': 3, 'throughput': 1.2705703520892857, 'latency_mean': 2.3532765114307406, 'latency_p50': 2.3495798110961914, 'latency_p90': 2.5827368021011354}, {'batch_size': 5, 'throughput': 1.561791576128465, 'latency_mean': 3.1910104477405548, 'latency_p50': 3.193638324737549, 'latency_p90': 3.6716687202453615}, {'batch_size': 6, 'throughput': 1.6310491545396348, 'latency_mean': 3.6608028805255888, 'latency_p50': 3.66618275642395, 'latency_p90': 4.149280834197998}, {'batch_size': 8, 'throughput': 1.753123955716314, 'latency_mean': 4.54364466547966, 'latency_p50': 4.528327107429504, 'latency_p90': 5.225707459449768}, {'batch_size': 10, 'throughput': 1.8335636320224424, 'latency_mean': 5.4211284697055815, 'latency_p50': 5.411407113075256, 'latency_p90': 6.127089762687683}]

gpu_counts: {'NVIDIA L40S': 1}

display_name: training123

is_internal_developer: True

language_model: ChaiML/ca18-v1-dpo_cosine

model_size: 13B

ranking_group: single

throughput_3p7s: 1.65

us_pacific_date: 2025-11-15

win_ratio: 0.5180145152928979

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', '</s>', 'User:', 'You:'], '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-dpo-cosine-v1-mkmlizer
Waiting for job on chaiml-ca18-v1-dpo-cosine-v1-mkmlizer to finish
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: ║                                                                     ║
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: ║   ██████   ██████  █████   ████  ████                               ║
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: ║  ░░██████ ██████  ░░███   ███░  ░░███                               ║
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: ║   ░███░█████░███   ░███  ███     ░███                               ║
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: ║   ░███░░███ ░███   ░███████      ░███                               ║
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: ║   ░███ ░░░  ░███   ░███░░███     ░███                               ║
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: ║   ░███      ░███   ░███ ░░███    ░███                               ║
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: ║   █████     █████  █████ ░░████  █████                              ║
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: ║  ░░░░░     ░░░░░  ░░░░░   ░░░░  ░░░░░                               ║
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: ║                                                                     ║
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: ║  Version: 0.30.2                                                    ║
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: ║  Features: FLYWHEEL, CUDA                                           ║
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: ║  Copyright 2023-2025 MK ONE TECHNOLOGIES Inc.                       ║
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: ║  https://mk1.ai                                                     ║
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: ║                                                                     ║
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: ║  belonging to:                                                      ║
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: ║                                                                     ║
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: ║  Chai Research Corp.                                                ║
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: ║  Expiration: 2028-03-31 23:59:59                                    ║
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: ║                                                                     ║
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: Downloaded to shared memory in 22.618s
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: Checking if ChaiML/ca18-v1-dpo_cosine already exists in ChaiML
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp32_qcycq, device:0
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: quantized model in 25.914s
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: Processed model ChaiML/ca18-v1-dpo_cosine in 48.532s
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: creating bucket guanaco-mkml-models
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-ca18-v1-dpo-cosine-v1/nvidia
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-ca18-v1-dpo-cosine-v1/nvidia/config.json
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-ca18-v1-dpo-cosine-v1/nvidia/special_tokens_map.json
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-ca18-v1-dpo-cosine-v1/nvidia/tokenizer_config.json
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-ca18-v1-dpo-cosine-v1/nvidia/tokenizer.json
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-ca18-v1-dpo-cosine-v1/nvidia/flywheel_model.0.safetensors
chaiml-ca18-v1-dpo-cosine-v1-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|▏         | 5/363 [00:00<00:10, 34.24it/s]
Loading 0:   4%|▍         | 14/363 [00:00<00:07, 48.54it/s]
Loading 0:   6%|▋         | 23/363 [00:00<00:06, 52.61it/s]
Loading 0:   9%|▉         | 32/363 [00:00<00:06, 54.72it/s]
Loading 0:  11%|█▏        | 41/363 [00:00<00:05, 55.97it/s]
Loading 0:  14%|█▍        | 50/363 [00:00<00:05, 56.40it/s]
Loading 0:  17%|█▋        | 60/363 [00:01<00:04, 62.30it/s]
Loading 0:  18%|█▊        | 67/363 [00:01<00:06, 46.32it/s]
Loading 0:  20%|██        | 73/363 [00:01<00:06, 43.94it/s]
Loading 0:  23%|██▎       | 82/363 [00:01<00:05, 47.28it/s]
Loading 0:  25%|██▍       | 90/363 [00:01<00:05, 53.87it/s]
Loading 0:  26%|██▋       | 96/363 [00:01<00:05, 52.36it/s]
Loading 0:  28%|██▊       | 102/363 [00:01<00:04, 53.03it/s]
Loading 0:  30%|███       | 110/363 [00:02<00:04, 55.11it/s]
Loading 0:  32%|███▏      | 116/363 [00:02<00:04, 54.59it/s]
Loading 0:  34%|███▎      | 122/363 [00:02<00:04, 55.87it/s]
Loading 0:  35%|███▌      | 128/363 [00:02<00:04, 49.81it/s]
Loading 0:  37%|███▋      | 135/363 [00:02<00:04, 54.19it/s]
Loading 0:  39%|███▉      | 141/363 [00:02<00:04, 52.20it/s]
Loading 0:  40%|████      | 147/363 [00:02<00:05, 37.69it/s]
Loading 0:  42%|████▏     | 153/363 [00:03<00:04, 42.17it/s]
Loading 0:  44%|████▎     | 158/363 [00:03<00:05, 39.61it/s]
Loading 0:  46%|████▌     | 166/363 [00:03<00:04, 48.35it/s]
Loading 0:  47%|████▋     | 172/363 [00:03<00:04, 47.21it/s]
Loading 0:  49%|████▉     | 178/363 [00:03<00:03, 48.97it/s]
Loading 0:  51%|█████     | 185/363 [00:03<00:03, 47.06it/s]
Loading 0:  53%|█████▎    | 193/363 [00:03<00:03, 54.41it/s]
Loading 0:  55%|█████▍    | 199/363 [00:03<00:03, 52.30it/s]
Loading 0:  56%|█████▋    | 205/363 [00:04<00:02, 52.71it/s]
Loading 0:  58%|█████▊    | 212/363 [00:04<00:03, 49.31it/s]
Loading 0:  61%|██████    | 222/363 [00:04<00:02, 55.54it/s]
Loading 0:  63%|██████▎   | 228/363 [00:04<00:03, 42.44it/s]
Loading 0:  64%|██████▍   | 234/363 [00:04<00:02, 45.96it/s]
Loading 0:  66%|██████▌   | 240/363 [00:04<00:02, 44.43it/s]
Loading 0:  68%|██████▊   | 247/363 [00:04<00:02, 50.17it/s]
Loading 0:  70%|██████▉   | 253/363 [00:05<00:02, 49.25it/s]
Loading 0:  71%|███████▏  | 259/363 [00:05<00:02, 51.10it/s]
Loading 0:  73%|███████▎  | 266/363 [00:05<00:02, 48.26it/s]
Loading 0:  75%|███████▌  | 274/363 [00:05<00:01, 55.29it/s]
Loading 0:  77%|███████▋  | 280/363 [00:05<00:01, 53.40it/s]
Loading 0:  79%|███████▉  | 286/363 [00:05<00:01, 53.59it/s]
Loading 0:  81%|████████  | 293/363 [00:05<00:01, 49.49it/s]
Loading 0:  83%|████████▎ | 303/363 [00:06<00:01, 54.86it/s]
Loading 0:  85%|████████▌ | 309/363 [00:06<00:01, 33.34it/s]
Loading 0:  87%|████████▋ | 314/363 [00:06<00:01, 34.76it/s]
Loading 0:  88%|████████▊ | 321/363 [00:06<00:01, 37.56it/s]
Loading 0:  91%|█████████ | 330/363 [00:06<00:00, 43.06it/s]
Loading 0:  93%|█████████▎| 339/363 [00:07<00:00, 46.67it/s]
Loading 0:  96%|█████████▌| 347/363 [00:07<00:00, 53.40it/s]
Loading 0:  97%|█████████▋| 353/363 [00:07<00:00, 51.87it/s]
Loading 0:  99%|█████████▉| 359/363 [00:07<00:00, 52.79it/s]
                                                            
Job chaiml-ca18-v1-dpo-cosine-v1-mkmlizer completed after 93.27s with status: succeeded
Stopping job with name chaiml-ca18-v1-dpo-cosine-v1-mkmlizer
Pipeline stage MKMLizer completed in 94.02s
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-ca18-v1-dpo-cosine-v1
Waiting for inference service chaiml-ca18-v1-dpo-cosine-v1 to be ready
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
Inference service chaiml-ca18-v1-dpo-cosine-v1 ready after 170.85456347465515s
Pipeline stage MKMLDeployer completed in 171.44s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 1.8963465690612793s
Received healthy response to inference request in 1.6261491775512695s
Received healthy response to inference request in 1.1855659484863281s
Received healthy response to inference request in 1.500519037246704s
Received healthy response to inference request in 1.462634801864624s
5 requests
0 failed requests
5th percentile: 1.2409797191619873
10th percentile: 1.2963934898376466
20th percentile: 1.4072210311889648
30th percentile: 1.47021164894104
40th percentile: 1.485365343093872
50th percentile: 1.500519037246704
60th percentile: 1.5507710933685304
70th percentile: 1.6010231494903564
80th percentile: 1.6801886558532715
90th percentile: 1.7882676124572754
95th percentile: 1.8423070907592773
99th percentile: 1.885538673400879
mean time: 1.534243106842041
Pipeline stage StressChecker completed in 8.88s
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 0.62s
Shutdown handler de-registered
chaiml-ca18-v1-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 0.08s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service chaiml-ca18-v1-dpo-cosine-v1-profiler
Waiting for inference service chaiml-ca18-v1-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 2237.27s
Shutdown handler de-registered
chaiml-ca18-v1-dpo-cosine_v1 status is now inactive due to auto deactivation removed underperforming models
chaiml-ca18-v1-dpo-cosine_v1 status is now torndown due to DeploymentManager action