chaiml-ca18-v1-3-dpo

developer_uid: chai_backend_admin

submission_id: chaiml-ca18-v1-3-dpo_v1

model_name: training123

model_group: ChaiML/ca18-v1-3-dpo

status: torndown

timestamp: 2025-11-19T03:26:33+00:00

num_battles: 7363

num_wins: 3999

celo_rating: 1316.17

family_friendly_score: 0.5085999999999999

family_friendly_standard_error: 0.0070700217821446625

submission_type: basic

model_repo: ChaiML/ca18-v1-3-dpo

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': 0.6830221188701149, 'latency_mean': 1.4639598190784455, 'latency_p50': 1.4627426862716675, 'latency_p90': 1.599713635444641}, {'batch_size': 3, 'throughput': 1.2825697444183541, 'latency_mean': 2.3292975211143494, 'latency_p50': 2.3277734518051147, 'latency_p90': 2.576637053489685}, {'batch_size': 5, 'throughput': 1.5717255304353137, 'latency_mean': 3.1717317414283754, 'latency_p50': 3.1640799045562744, 'latency_p90': 3.600573968887329}, {'batch_size': 6, 'throughput': 1.6503954220578005, 'latency_mean': 3.6051815056800844, 'latency_p50': 3.6126970052719116, 'latency_p90': 4.1374413013458256}, {'batch_size': 8, 'throughput': 1.7839074569085194, 'latency_mean': 4.46465092420578, 'latency_p50': 4.465772986412048, 'latency_p90': 5.0662551641464235}, {'batch_size': 10, 'throughput': 1.8519249378798504, 'latency_mean': 5.360911712646485, 'latency_p50': 5.380105257034302, 'latency_p90': 6.010298585891723}]

gpu_counts: {'NVIDIA L40S': 1}

display_name: training123

is_internal_developer: True

language_model: ChaiML/ca18-v1-3-dpo

model_size: 13B

ranking_group: single

throughput_3p7s: 1.68

us_pacific_date: 2025-11-15

win_ratio: 0.5431210104576939

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': ['You:', '\n', '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
Connection pool is full, discarding connection: %s. Connection pool size: %s
run pipeline %s
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
run pipeline stage %s
Running pipeline stage MKMLizer
Starting job with name chaiml-ca18-v1-3-dpo-v1-mkmlizer
Waiting for job on chaiml-ca18-v1-3-dpo-v1-mkmlizer to finish
chaiml-ca18-v1-3-dpo-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-ca18-v1-3-dpo-v1-mkmlizer: ║                                                                     ║
chaiml-ca18-v1-3-dpo-v1-mkmlizer: ║   ██████   ██████  █████   ████  ████                               ║
chaiml-ca18-v1-3-dpo-v1-mkmlizer: ║  ░░██████ ██████  ░░███   ███░  ░░███                               ║
chaiml-ca18-v1-3-dpo-v1-mkmlizer: ║   ░███░█████░███   ░███  ███     ░███                               ║
chaiml-ca18-v1-3-dpo-v1-mkmlizer: ║   ░███░░███ ░███   ░███████      ░███                               ║
chaiml-ca18-v1-3-dpo-v1-mkmlizer: ║   ░███ ░░░  ░███   ░███░░███     ░███                               ║
chaiml-ca18-v1-3-dpo-v1-mkmlizer: ║   ░███      ░███   ░███ ░░███    ░███                               ║
chaiml-ca18-v1-3-dpo-v1-mkmlizer: ║   █████     █████  █████ ░░████  █████                              ║
chaiml-ca18-v1-3-dpo-v1-mkmlizer: ║  ░░░░░     ░░░░░  ░░░░░   ░░░░  ░░░░░                               ║
chaiml-ca18-v1-3-dpo-v1-mkmlizer: ║                                                                     ║
chaiml-ca18-v1-3-dpo-v1-mkmlizer: ║  Version: 0.30.2                                                    ║
chaiml-ca18-v1-3-dpo-v1-mkmlizer: ║  Features: FLYWHEEL, CUDA                                           ║
chaiml-ca18-v1-3-dpo-v1-mkmlizer: ║  Copyright 2023-2025 MK ONE TECHNOLOGIES Inc.                       ║
chaiml-ca18-v1-3-dpo-v1-mkmlizer: ║  https://mk1.ai                                                     ║
chaiml-ca18-v1-3-dpo-v1-mkmlizer: ║                                                                     ║
chaiml-ca18-v1-3-dpo-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
chaiml-ca18-v1-3-dpo-v1-mkmlizer: ║  belonging to:                                                      ║
chaiml-ca18-v1-3-dpo-v1-mkmlizer: ║                                                                     ║
chaiml-ca18-v1-3-dpo-v1-mkmlizer: ║  Chai Research Corp.                                                ║
chaiml-ca18-v1-3-dpo-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
chaiml-ca18-v1-3-dpo-v1-mkmlizer: ║  Expiration: 2028-03-31 23:59:59                                    ║
chaiml-ca18-v1-3-dpo-v1-mkmlizer: ║                                                                     ║
chaiml-ca18-v1-3-dpo-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
chaiml-ca18-v1-3-dpo-v1-mkmlizer: Downloaded to shared memory in 84.524s
chaiml-ca18-v1-3-dpo-v1-mkmlizer: Checking if ChaiML/ca18-v1-3-dpo already exists in ChaiML
chaiml-ca18-v1-3-dpo-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp_dt1sen0, device:0
chaiml-ca18-v1-3-dpo-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
chaiml-ca18-v1-3-dpo-v1-mkmlizer: quantized model in 26.352s
chaiml-ca18-v1-3-dpo-v1-mkmlizer: Processed model ChaiML/ca18-v1-3-dpo in 110.877s
chaiml-ca18-v1-3-dpo-v1-mkmlizer: creating bucket guanaco-mkml-models
chaiml-ca18-v1-3-dpo-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-ca18-v1-3-dpo-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-ca18-v1-3-dpo-v1/nvidia
chaiml-ca18-v1-3-dpo-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-ca18-v1-3-dpo-v1/nvidia/config.json
chaiml-ca18-v1-3-dpo-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-ca18-v1-3-dpo-v1/nvidia/special_tokens_map.json
chaiml-ca18-v1-3-dpo-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-ca18-v1-3-dpo-v1/nvidia/tokenizer_config.json
chaiml-ca18-v1-3-dpo-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-ca18-v1-3-dpo-v1/nvidia/tokenizer.json
chaiml-ca18-v1-3-dpo-v1-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|▏         | 5/363 [00:00<00:10, 33.13it/s]
Loading 0:   4%|▍         | 14/363 [00:00<00:07, 46.59it/s]
Loading 0:   6%|▋         | 23/363 [00:00<00:06, 50.70it/s]
Loading 0:   9%|▉         | 32/363 [00:00<00:06, 52.88it/s]
Loading 0:  11%|█▏        | 41/363 [00:00<00:05, 54.05it/s]
Loading 0:  14%|█▍        | 50/363 [00:00<00:05, 55.08it/s]
Loading 0:  17%|█▋        | 60/363 [00:01<00:04, 60.83it/s]
Loading 0:  18%|█▊        | 67/363 [00:01<00:06, 44.01it/s]
Loading 0:  20%|██        | 73/363 [00:01<00:06, 42.55it/s]
Loading 0:  22%|██▏       | 81/363 [00:01<00:05, 49.71it/s]
Loading 0:  24%|██▍       | 87/363 [00:01<00:05, 48.95it/s]
Loading 0:  26%|██▌       | 93/363 [00:01<00:05, 50.61it/s]
Loading 0:  28%|██▊       | 100/363 [00:02<00:05, 48.41it/s]
Loading 0:  30%|███       | 110/363 [00:02<00:04, 54.13it/s]
Loading 0:  32%|███▏      | 116/363 [00:02<00:04, 54.56it/s]
Loading 0:  34%|███▎      | 122/363 [00:02<00:04, 55.41it/s]
Loading 0:  35%|███▌      | 128/363 [00:02<00:04, 49.55it/s]
Loading 0:  37%|███▋      | 135/363 [00:02<00:04, 53.51it/s]
Loading 0:  39%|███▉      | 141/363 [00:02<00:04, 50.94it/s]
Loading 0:  40%|████      | 147/363 [00:03<00:05, 38.20it/s]
Loading 0:  42%|████▏     | 153/363 [00:03<00:04, 42.35it/s]
Loading 0:  44%|████▎     | 158/363 [00:03<00:05, 38.86it/s]
Loading 0:  46%|████▌     | 166/363 [00:03<00:04, 47.11it/s]
Loading 0:  47%|████▋     | 172/363 [00:03<00:04, 46.69it/s]
Loading 0:  49%|████▉     | 178/363 [00:03<00:03, 48.68it/s]
Loading 0:  51%|█████     | 185/363 [00:03<00:03, 46.25it/s]
Loading 0:  53%|█████▎    | 193/363 [00:03<00:03, 53.43it/s]
Loading 0:  55%|█████▍    | 199/363 [00:04<00:03, 51.83it/s]
Loading 0:  56%|█████▋    | 205/363 [00:04<00:02, 52.72it/s]
Loading 0:  58%|█████▊    | 212/363 [00:04<00:03, 49.62it/s]
Loading 0:  61%|██████    | 222/363 [00:04<00:02, 55.84it/s]
Loading 0:  63%|██████▎   | 228/363 [00:04<00:03, 42.09it/s]
Loading 0:  64%|██████▍   | 234/363 [00:04<00:02, 45.36it/s]
Loading 0:  66%|██████▌   | 240/363 [00:04<00:02, 42.86it/s]
Loading 0:  68%|██████▊   | 247/363 [00:05<00:02, 48.11it/s]
Loading 0:  70%|██████▉   | 253/363 [00:05<00:02, 47.82it/s]
Loading 0:  71%|███████▏  | 259/363 [00:05<00:02, 49.68it/s]
Loading 0:  73%|███████▎  | 266/363 [00:05<00:02, 47.44it/s]
Loading 0:  75%|███████▌  | 274/363 [00:05<00:01, 54.42it/s]
Loading 0:  77%|███████▋  | 280/363 [00:05<00:01, 50.74it/s]
Loading 0:  79%|███████▉  | 286/363 [00:05<00:01, 51.67it/s]
Loading 0:  81%|████████  | 293/363 [00:06<00:01, 47.61it/s]
Loading 0:  83%|████████▎ | 303/363 [00:06<00:01, 53.89it/s]
Loading 0:  85%|████████▌ | 309/363 [00:06<00:01, 32.18it/s]
Loading 0:  87%|████████▋ | 314/363 [00:06<00:01, 33.52it/s]
Loading 0:  88%|████████▊ | 321/363 [00:06<00:01, 36.58it/s]
Loading 0:  91%|█████████ | 330/363 [00:06<00:00, 42.11it/s]
Loading 0:  93%|█████████▎| 339/363 [00:07<00:00, 46.14it/s]
Loading 0:  96%|█████████▌| 347/363 [00:07<00:00, 52.93it/s]
Loading 0:  97%|█████████▋| 353/363 [00:07<00:00, 51.43it/s]
Loading 0:  99%|█████████▉| 359/363 [00:07<00:00, 52.31it/s]
                                                            
Job chaiml-ca18-v1-3-dpo-v1-mkmlizer completed after 159.47s with status: succeeded
Stopping job with name chaiml-ca18-v1-3-dpo-v1-mkmlizer
Pipeline stage MKMLizer completed in 160.70s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 1.16s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service chaiml-ca18-v1-3-dpo-v1
Waiting for inference service chaiml-ca18-v1-3-dpo-v1 to be ready
Failed to get response for submission mistralai-mistral-nem_93303_v569: ('http://mistralai-mistral-nem-93303-v569-predictor.tenant-chaiml-guanaco.k2.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', '')
Inference service chaiml-ca18-v1-3-dpo-v1 ready after 160.83998036384583s
Pipeline stage MKMLDeployer completed in 162.72s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 1.9057636260986328s
Received healthy response to inference request in 1.2931358814239502s
Received healthy response to inference request in 1.2686967849731445s
Received healthy response to inference request in 1.6119356155395508s
Received healthy response to inference request in 1.2482717037200928s
5 requests
0 failed requests
5th percentile: 1.252356719970703
10th percentile: 1.2564417362213134
20th percentile: 1.2646117687225342
30th percentile: 1.2735846042633057
40th percentile: 1.283360242843628
50th percentile: 1.2931358814239502
60th percentile: 1.4206557750701905
70th percentile: 1.5481756687164305
80th percentile: 1.6707012176513671
90th percentile: 1.788232421875
95th percentile: 1.8469980239868165
99th percentile: 1.8940105056762695
mean time: 1.4655607223510743
Pipeline stage StressChecker completed in 8.57s
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 0.59s
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 0.98s
Shutdown handler de-registered
chaiml-ca18-v1-3-dpo_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 2560.85s
Shutdown handler de-registered
chaiml-ca18-v1-3-dpo_v1 status is now inactive due to auto deactivation removed underperforming models
chaiml-ca18-v1-3-dpo_v1 status is now torndown due to DeploymentManager action