chaiml-ir-v3-3-dpo-qcld

developer_uid: chai_backend_admin

submission_id: chaiml-ir-v3-3-dpo-qcld_v1

model_name: training123

model_group: ChaiML/ir-v3-3-dpo_qcld

status: torndown

timestamp: 2025-11-17T13:31:26+00:00

num_battles: 5785

num_wins: 2997

celo_rating: 1297.33

family_friendly_score: 0.5096

family_friendly_standard_error: 0.007069764352508505

submission_type: basic

model_repo: ChaiML/ir-v3-3-dpo_qcld

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': 0.3755060670869137, 'latency_mean': 2.6629747986793517, 'latency_p50': 2.660498261451721, 'latency_p90': 2.9270139932632446}, {'batch_size': 2, 'throughput': 0.551860697787945, 'latency_mean': 3.6186133670806884, 'latency_p50': 3.620659351348877, 'latency_p90': 3.913903427124023}, {'batch_size': 3, 'throughput': 0.6769716771407407, 'latency_mean': 4.4220716869831085, 'latency_p50': 4.416670560836792, 'latency_p90': 4.774251008033752}, {'batch_size': 4, 'throughput': 0.767657937030225, 'latency_mean': 5.1955054974555965, 'latency_p50': 5.195307612419128, 'latency_p90': 5.9327953338623045}, {'batch_size': 5, 'throughput': 0.8170308548896206, 'latency_mean': 6.102747325897217, 'latency_p50': 6.0938193798065186, 'latency_p90': 7.062655329704285}]

gpu_counts: {'NVIDIA L40S': 1}

display_name: training123

is_internal_developer: True

language_model: ChaiML/ir-v3-3-dpo_qcld

model_size: 24B

ranking_group: single

throughput_3p7s: 0.57

us_pacific_date: 2025-11-14

win_ratio: 0.5180639585133967

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', '</s>', '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
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
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
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 %s
run pipeline stage %s
Running pipeline stage MKMLizer
Starting job with name chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer
Waiting for job on chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer to finish
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: ║                                                                     ║
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: ║   ██████   ██████  █████   ████  ████                               ║
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: ║  ░░██████ ██████  ░░███   ███░  ░░███                               ║
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: ║   ░███░█████░███   ░███  ███     ░███                               ║
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: ║   ░███░░███ ░███   ░███████      ░███                               ║
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: ║   ░███ ░░░  ░███   ░███░░███     ░███                               ║
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: ║   ░███      ░███   ░███ ░░███    ░███                               ║
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: ║   █████     █████  █████ ░░████  █████                              ║
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: ║  ░░░░░     ░░░░░  ░░░░░   ░░░░  ░░░░░                               ║
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: ║                                                                     ║
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: ║  Version: 0.30.2                                                    ║
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: ║  Features: FLYWHEEL, CUDA                                           ║
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: ║  Copyright 2023-2025 MK ONE TECHNOLOGIES Inc.                       ║
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: ║  https://mk1.ai                                                     ║
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: ║                                                                     ║
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: ║  belonging to:                                                      ║
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: ║                                                                     ║
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: ║  Chai Research Corp.                                                ║
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: ║  Expiration: 2028-03-31 23:59:59                                    ║
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: ║                                                                     ║
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: Downloaded to shared memory in 50.657s
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: Checking if ChaiML/ir-v3-3-dpo_qcld already exists in ChaiML
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpcdi2gomb, device:0
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: quantized model in 43.520s
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: Processed model ChaiML/ir-v3-3-dpo_qcld in 94.178s
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: creating bucket guanaco-mkml-models
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-ir-v3-3-dpo-qcld-v1/nvidia
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-ir-v3-3-dpo-qcld-v1/nvidia/config.json
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-ir-v3-3-dpo-qcld-v1/nvidia/special_tokens_map.json
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-ir-v3-3-dpo-qcld-v1/nvidia/tokenizer_config.json
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-ir-v3-3-dpo-qcld-v1/nvidia/tokenizer.json
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', '')
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-ir-v3-3-dpo-qcld-v1/nvidia/flywheel_model.0.safetensors
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/chaiml-ir-v3-3-dpo-qcld-v1/nvidia/flywheel_model.1.safetensors
chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|▏         | 5/363 [00:00<00:15, 22.95it/s]
Loading 0:   3%|▎         | 12/363 [00:00<00:08, 41.45it/s]
Loading 0:   5%|▍         | 18/363 [00:00<00:08, 39.33it/s]
Loading 0:   6%|▋         | 23/363 [00:00<00:10, 31.21it/s]
Loading 0:   9%|▉         | 32/363 [00:00<00:07, 43.22it/s]
Loading 0:  10%|█         | 38/363 [00:01<00:10, 30.36it/s]
Loading 0:  12%|█▏        | 42/363 [00:01<00:11, 27.25it/s]
Loading 0:  13%|█▎        | 48/363 [00:01<00:09, 32.33it/s]
Loading 0:  14%|█▍        | 52/363 [00:01<00:09, 31.13it/s]
Loading 0:  16%|█▌        | 57/363 [00:01<00:08, 34.64it/s]
Loading 0:  17%|█▋        | 61/363 [00:01<00:09, 32.50it/s]
Loading 0:  18%|█▊        | 65/363 [00:01<00:08, 33.54it/s]
Loading 0:  19%|█▉        | 70/363 [00:02<00:09, 30.72it/s]
Loading 0:  20%|██        | 74/363 [00:02<00:10, 27.04it/s]
Loading 0:  22%|██▏       | 79/363 [00:02<00:09, 30.94it/s]
Loading 0:  23%|██▎       | 83/363 [00:02<00:08, 31.87it/s]
Loading 0:  24%|██▍       | 87/363 [00:02<00:09, 28.11it/s]
Loading 0:  25%|██▌       | 92/363 [00:02<00:10, 26.94it/s]
Loading 0:  27%|██▋       | 99/363 [00:03<00:07, 34.17it/s]
Loading 0:  28%|██▊       | 103/363 [00:03<00:08, 32.27it/s]
Loading 0:  29%|██▉       | 107/363 [00:03<00:08, 28.72it/s]
Loading 0:  31%|███       | 112/363 [00:03<00:08, 31.15it/s]
Loading 0:  32%|███▏      | 116/363 [00:03<00:07, 31.55it/s]
Loading 0:  33%|███▎      | 120/363 [00:03<00:07, 33.25it/s]
Loading 0:  34%|███▍      | 124/363 [00:03<00:07, 31.47it/s]
Loading 0:  36%|███▌      | 129/363 [00:04<00:06, 34.69it/s]
Loading 0:  37%|███▋      | 133/363 [00:04<00:07, 32.23it/s]
Loading 0:  38%|███▊      | 138/363 [00:04<00:06, 35.37it/s]
Loading 0:  39%|███▉      | 142/363 [00:04<00:06, 32.82it/s]
Loading 0:  41%|████      | 149/363 [00:04<00:05, 39.03it/s]
Loading 0:  42%|████▏     | 153/363 [00:04<00:08, 25.14it/s]
Loading 0:  44%|████▎     | 158/363 [00:05<00:08, 25.27it/s]
Loading 0:  45%|████▌     | 165/363 [00:05<00:06, 31.84it/s]
Loading 0:  47%|████▋     | 169/363 [00:05<00:06, 30.55it/s]
Loading 0:  48%|████▊     | 174/363 [00:05<00:05, 33.64it/s]
Loading 0:  49%|████▉     | 178/363 [00:05<00:05, 31.71it/s]
Loading 0:  50%|█████     | 182/363 [00:05<00:05, 32.49it/s]
Loading 0:  52%|█████▏    | 187/363 [00:05<00:05, 29.42it/s]
Loading 0:  53%|█████▎    | 191/363 [00:06<00:05, 28.77it/s]
Loading 0:  54%|█████▎    | 195/363 [00:06<00:06, 25.35it/s]
Loading 0:  55%|█████▌    | 201/363 [00:19<02:13,  1.21it/s]
Loading 0:  56%|█████▌    | 203/363 [00:19<01:55,  1.38it/s]
Loading 0:  57%|█████▋    | 208/363 [00:19<01:14,  2.08it/s]
Loading 0:  58%|█████▊    | 211/363 [00:19<00:57,  2.63it/s]
Loading 0:  59%|█████▉    | 214/363 [00:19<00:43,  3.40it/s]
Loading 0:  60%|██████    | 218/363 [00:20<00:30,  4.70it/s]
Loading 0:  61%|██████    | 221/363 [00:20<00:25,  5.65it/s]
Loading 0:  62%|██████▏   | 224/363 [00:20<00:20,  6.83it/s]
Loading 0:  63%|██████▎   | 227/363 [00:20<00:16,  8.09it/s]
Loading 0:  63%|██████▎   | 229/363 [00:20<00:15,  8.41it/s]
Loading 0:  64%|██████▎   | 231/363 [00:20<00:14,  9.33it/s]
Loading 0:  65%|██████▍   | 235/363 [00:21<00:09, 13.21it/s]
Loading 0:  66%|██████▌   | 238/363 [00:21<00:07, 15.66it/s]
Loading 0:  66%|██████▋   | 241/363 [00:21<00:07, 16.52it/s]
Loading 0:  67%|██████▋   | 244/363 [00:21<00:06, 17.46it/s]
Loading 0:  68%|██████▊   | 247/363 [00:21<00:05, 19.95it/s]
Loading 0:  69%|██████▉   | 250/363 [00:21<00:05, 21.56it/s]
Loading 0:  70%|██████▉   | 253/363 [00:21<00:05, 21.17it/s]
Loading 0:  71%|███████   | 257/363 [00:22<00:05, 20.96it/s]
Loading 0:  73%|███████▎  | 266/363 [00:22<00:02, 33.39it/s]
Loading 0:  74%|███████▍  | 270/363 [00:22<00:04, 21.84it/s]
Loading 0:  76%|███████▌  | 275/363 [00:22<00:03, 22.96it/s]
Loading 0:  78%|███████▊  | 282/363 [00:22<00:02, 29.84it/s]
Loading 0:  79%|███████▉  | 286/363 [00:23<00:02, 29.52it/s]
Loading 0:  80%|███████▉  | 290/363 [00:23<00:02, 28.73it/s]
Loading 0:  81%|████████  | 294/363 [00:23<00:03, 20.50it/s]
Loading 0:  82%|████████▏ | 298/363 [00:23<00:02, 21.89it/s]
Loading 0:  83%|████████▎ | 301/363 [00:23<00:02, 22.97it/s]
Loading 0:  84%|████████▎ | 304/363 [00:23<00:02, 20.66it/s]
Loading 0:  85%|████████▍ | 307/363 [00:24<00:02, 20.66it/s]
Loading 0:  86%|████████▌ | 311/363 [00:24<00:02, 20.90it/s]
Loading 0:  88%|████████▊ | 318/363 [00:24<00:01, 29.41it/s]
Loading 0:  89%|████████▊ | 322/363 [00:24<00:01, 28.96it/s]
Loading 0:  90%|█████████ | 327/363 [00:24<00:01, 32.75it/s]
Loading 0:  91%|█████████ | 331/363 [00:24<00:01, 31.37it/s]
Loading 0:  92%|█████████▏| 335/363 [00:24<00:00, 32.41it/s]
Loading 0:  93%|█████████▎| 339/363 [00:25<00:00, 32.96it/s]
Loading 0:  94%|█████████▍| 343/363 [00:25<00:00, 20.27it/s]
Loading 0:  96%|█████████▌| 348/363 [00:25<00:00, 21.85it/s]
Loading 0:  98%|█████████▊| 355/363 [00:25<00:00, 29.13it/s]
Loading 0:  99%|█████████▉| 359/363 [00:25<00:00, 28.95it/s]
                                                            
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', '')
Job chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer completed after 175.34s with status: succeeded
Stopping job with name chaiml-ir-v3-3-dpo-qcld-v1-mkmlizer
Pipeline stage MKMLizer completed in 175.83s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.16s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service chaiml-ir-v3-3-dpo-qcld-v1
Waiting for inference service chaiml-ir-v3-3-dpo-qcld-v1 to be ready
Inference service chaiml-ir-v3-3-dpo-qcld-v1 ready after 150.65721583366394s
Pipeline stage MKMLDeployer completed in 151.21s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.48482084274292s
Received healthy response to inference request in 3.3018107414245605s
Received healthy response to inference request in 2.42203950881958s
Received healthy response to inference request in 2.5104727745056152s
Received healthy response to inference request in 2.538206100463867s
5 requests
0 failed requests
5th percentile: 2.4345957756042482
10th percentile: 2.447152042388916
20th percentile: 2.4722645759582518
30th percentile: 2.489951229095459
40th percentile: 2.500212001800537
50th percentile: 2.5104727745056152
60th percentile: 2.521566104888916
70th percentile: 2.5326594352722167
80th percentile: 2.690927028656006
90th percentile: 2.996368885040283
95th percentile: 3.149089813232422
99th percentile: 3.271266555786133
mean time: 2.6514699935913084
Pipeline stage StressChecker completed in 14.56s
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 0.70s
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 0.81s
Shutdown handler de-registered
chaiml-ir-v3-3-dpo-qcld_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 3234.87s
Shutdown handler de-registered
chaiml-ir-v3-3-dpo-qcld_v1 status is now inactive due to auto deactivation removed underperforming models
chaiml-ir-v3-3-dpo-qcld_v1 status is now torndown due to DeploymentManager action