chaiml-ir-v3-4-dpo-irv0f_3260

developer_uid: chai_backend_admin

submission_id: chaiml-ir-v3-4-dpo-irv0f_3260_v1

model_name: training123

model_group: ChaiML/ir-v3-4-dpo_irv0f

status: torndown

timestamp: 2025-11-17T13:16:31+00:00

num_battles: 6383

num_wins: 3251

celo_rating: 1292.85

family_friendly_score: 0.0

family_friendly_standard_error: 0.0

submission_type: basic

model_repo: ChaiML/ir-v3-4-dpo_irv0fullpyg

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': 0.37371466797282804, 'latency_mean': 2.6757277250289917, 'latency_p50': 2.678100109100342, 'latency_p90': 2.929746222496033}, {'batch_size': 2, 'throughput': 0.5523286658965282, 'latency_mean': 3.613423852920532, 'latency_p50': 3.626803398132324, 'latency_p90': 3.920777177810669}, {'batch_size': 3, 'throughput': 0.6748305815129987, 'latency_mean': 4.427009390592575, 'latency_p50': 4.407195448875427, 'latency_p90': 5.001068019866943}, {'batch_size': 4, 'throughput': 0.7699463443751405, 'latency_mean': 5.180333551168442, 'latency_p50': 5.182057738304138, 'latency_p90': 5.78854455947876}, {'batch_size': 5, 'throughput': 0.8241961790421054, 'latency_mean': 6.051288194656372, 'latency_p50': 6.050435781478882, 'latency_p90': 6.912726092338562}]

gpu_counts: {'NVIDIA L40S': 1}

display_name: training123

is_internal_developer: True

language_model: ChaiML/ir-v3-4-dpo_irv0fullpyg

model_size: 24B

ranking_group: single

throughput_3p7s: 0.57

us_pacific_date: 2025-11-14

win_ratio: 0.509321635594548

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': ['User:', 'You:', '\n', '</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-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer
Waiting for job on chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer to finish
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: ║                                                                     ║
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: ║   ██████   ██████  █████   ████  ████                               ║
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: ║  ░░██████ ██████  ░░███   ███░  ░░███                               ║
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: ║   ░███░█████░███   ░███  ███     ░███                               ║
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: ║   ░███░░███ ░███   ░███████      ░███                               ║
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: ║   ░███ ░░░  ░███   ░███░░███     ░███                               ║
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: ║   ░███      ░███   ░███ ░░███    ░███                               ║
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: ║   █████     █████  █████ ░░████  █████                              ║
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: ║  ░░░░░     ░░░░░  ░░░░░   ░░░░  ░░░░░                               ║
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: ║                                                                     ║
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: ║  Version: 0.30.2                                                    ║
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: ║  Features: FLYWHEEL, CUDA                                           ║
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: ║  Copyright 2023-2025 MK ONE TECHNOLOGIES Inc.                       ║
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: ║  https://mk1.ai                                                     ║
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: ║                                                                     ║
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: ║  belonging to:                                                      ║
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: ║                                                                     ║
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: ║  Chai Research Corp.                                                ║
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: ║  Expiration: 2028-03-31 23:59:59                                    ║
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: ║                                                                     ║
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: Downloaded to shared memory in 65.535s
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: Checking if ChaiML/ir-v3-4-dpo_irv0fullpyg already exists in ChaiML
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp_ch2vnmr, device:0
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: quantized model in 42.206s
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: Processed model ChaiML/ir-v3-4-dpo_irv0fullpyg in 107.742s
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: creating bucket guanaco-mkml-models
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-ir-v3-4-dpo-irv0f-3260-v1/nvidia
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-ir-v3-4-dpo-irv0f-3260-v1/nvidia/config.json
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-ir-v3-4-dpo-irv0f-3260-v1/nvidia/special_tokens_map.json
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-ir-v3-4-dpo-irv0f-3260-v1/nvidia/tokenizer_config.json
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-ir-v3-4-dpo-irv0f-3260-v1/nvidia/tokenizer.json
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/chaiml-ir-v3-4-dpo-irv0f-3260-v1/nvidia/flywheel_model.1.safetensors
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-ir-v3-4-dpo-irv0f-3260-v1/nvidia/flywheel_model.0.safetensors
chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|▏         | 5/363 [00:00<00:15, 22.54it/s]
Loading 0:   4%|▎         | 13/363 [00:00<00:09, 38.84it/s]
Loading 0:   5%|▌         | 19/363 [00:00<00:08, 40.16it/s]
Loading 0:   7%|▋         | 24/363 [00:00<00:09, 35.72it/s]
Loading 0:   9%|▉         | 32/363 [00:00<00:07, 44.69it/s]
Loading 0:  10%|█         | 37/363 [00:01<00:11, 29.19it/s]
Loading 0:  11%|█▏        | 41/363 [00:01<00:12, 26.66it/s]
Loading 0:  13%|█▎        | 48/363 [00:01<00:09, 33.87it/s]
Loading 0:  15%|█▍        | 53/363 [00:01<00:09, 33.66it/s]
Loading 0:  16%|█▌        | 58/363 [00:01<00:08, 34.49it/s]
Loading 0:  17%|█▋        | 62/363 [00:01<00:08, 34.94it/s]
Loading 0:  18%|█▊        | 66/363 [00:01<00:08, 36.13it/s]
Loading 0:  19%|█▉        | 70/363 [00:02<00:10, 27.70it/s]
Loading 0:  20%|██        | 74/363 [00:02<00:11, 25.53it/s]
Loading 0:  22%|██▏       | 79/363 [00:02<00:09, 29.71it/s]
Loading 0:  23%|██▎       | 83/363 [00:02<00:09, 30.74it/s]
Loading 0:  24%|██▍       | 87/363 [00:02<00:10, 27.56it/s]
Loading 0:  25%|██▌       | 92/363 [00:02<00:10, 26.30it/s]
Loading 0:  27%|██▋       | 99/363 [00:03<00:07, 33.48it/s]
Loading 0:  28%|██▊       | 103/363 [00:03<00:07, 32.58it/s]
Loading 0:  29%|██▉       | 107/363 [00:03<00:09, 26.96it/s]
Loading 0:  31%|███       | 112/363 [00:03<00:09, 26.59it/s]
Loading 0:  32%|███▏      | 115/363 [00:03<00:09, 26.11it/s]
Loading 0:  33%|███▎      | 120/363 [00:03<00:08, 30.16it/s]
Loading 0:  34%|███▍      | 124/363 [00:04<00:08, 29.86it/s]
Loading 0:  36%|███▌      | 129/363 [00:04<00:06, 33.69it/s]
Loading 0:  37%|███▋      | 133/363 [00:04<00:07, 32.35it/s]
Loading 0:  38%|███▊      | 138/363 [00:04<00:06, 35.74it/s]
Loading 0:  39%|███▉      | 142/363 [00:04<00:07, 27.98it/s]
Loading 0:  41%|████      | 149/363 [00:04<00:06, 34.80it/s]
Loading 0:  42%|████▏     | 153/363 [00:05<00:08, 24.74it/s]
Loading 0:  44%|████▎     | 158/363 [00:05<00:08, 25.22it/s]
Loading 0:  45%|████▌     | 165/363 [00:05<00:06, 32.11it/s]
Loading 0:  47%|████▋     | 169/363 [00:05<00:06, 30.68it/s]
Loading 0:  48%|████▊     | 174/363 [00:05<00:05, 33.71it/s]
Loading 0:  49%|████▉     | 178/363 [00:05<00:05, 31.62it/s]
Loading 0:  50%|█████     | 182/363 [00:05<00:05, 32.71it/s]
Loading 0:  52%|█████▏    | 187/363 [00:06<00:05, 30.90it/s]
Loading 0:  53%|█████▎    | 191/363 [00:06<00:06, 27.10it/s]
Loading 0:  53%|█████▎    | 194/363 [00:06<00:07, 23.54it/s]
Loading 0:  55%|█████▌    | 201/363 [00:19<02:12,  1.22it/s]
Loading 0:  56%|█████▌    | 203/363 [00:19<01:54,  1.40it/s]
Loading 0:  57%|█████▋    | 208/363 [00:19<01:13,  2.11it/s]
Loading 0:  58%|█████▊    | 211/363 [00:19<00:57,  2.66it/s]
Loading 0:  59%|█████▉    | 214/363 [00:20<00:43,  3.41it/s]
Loading 0:  60%|██████    | 218/363 [00:20<00:30,  4.79it/s]
Loading 0:  61%|██████    | 221/363 [00:20<00:23,  6.09it/s]
Loading 0:  62%|██████▏   | 224/363 [00:20<00:18,  7.41it/s]
Loading 0:  63%|██████▎   | 229/363 [00:20<00:12, 10.78it/s]
Loading 0:  64%|██████▍   | 232/363 [00:20<00:10, 12.68it/s]
Loading 0:  65%|██████▌   | 237/363 [00:20<00:07, 17.29it/s]
Loading 0:  66%|██████▋   | 241/363 [00:20<00:06, 19.20it/s]
Loading 0:  68%|██████▊   | 246/363 [00:21<00:04, 23.89it/s]
Loading 0:  69%|██████▉   | 250/363 [00:21<00:04, 24.96it/s]
Loading 0:  70%|███████   | 255/363 [00:21<00:03, 29.26it/s]
Loading 0:  71%|███████▏  | 259/363 [00:21<00:03, 28.83it/s]
Loading 0:  73%|███████▎  | 266/363 [00:21<00:02, 35.86it/s]
Loading 0:  75%|███████▍  | 271/363 [00:21<00:03, 25.51it/s]
Loading 0:  76%|███████▌  | 275/363 [00:22<00:03, 24.10it/s]
Loading 0:  78%|███████▊  | 282/363 [00:22<00:02, 31.24it/s]
Loading 0:  79%|███████▉  | 286/363 [00:22<00:02, 30.46it/s]
Loading 0:  80%|████████  | 291/363 [00:22<00:02, 33.46it/s]
Loading 0:  81%|████████▏ | 295/363 [00:22<00:02, 31.59it/s]
Loading 0:  82%|████████▏ | 299/363 [00:22<00:01, 32.02it/s]
Loading 0:  84%|████████▎ | 304/363 [00:22<00:02, 29.20it/s]
Loading 0:  85%|████████▍ | 308/363 [00:23<00:01, 28.50it/s]
Loading 0:  86%|████████▌ | 311/363 [00:23<00:02, 23.80it/s]
Loading 0:  88%|████████▊ | 318/363 [00:23<00:01, 31.97it/s]
Loading 0:  89%|████████▊ | 322/363 [00:23<00:01, 30.85it/s]
Loading 0:  90%|█████████ | 327/363 [00:23<00:01, 34.10it/s]
Loading 0:  91%|█████████ | 331/363 [00:23<00:01, 32.00it/s]
Loading 0:  92%|█████████▏| 335/363 [00:23<00:00, 32.77it/s]
Loading 0:  93%|█████████▎| 339/363 [00:24<00:00, 33.59it/s]
Loading 0:  94%|█████████▍| 343/363 [00:24<00:00, 20.30it/s]
Loading 0:  96%|█████████▌| 348/363 [00:24<00:00, 22.11it/s]
Loading 0:  98%|█████████▊| 355/363 [00:24<00:00, 29.72it/s]
Loading 0:  99%|█████████▉| 359/363 [00:24<00:00, 29.23it/s]
                                                            
Job chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer completed after 154.8s with status: succeeded
Stopping job with name chaiml-ir-v3-4-dpo-irv0f-3260-v1-mkmlizer
Pipeline stage MKMLizer completed in 155.30s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.15s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service chaiml-ir-v3-4-dpo-irv0f-3260-v1
Waiting for inference service chaiml-ir-v3-4-dpo-irv0f-3260-v1 to be ready
Inference service chaiml-ir-v3-4-dpo-irv0f-3260-v1 ready after 161.31964588165283s
Pipeline stage MKMLDeployer completed in 161.94s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 3.147867202758789s
Received healthy response to inference request in 2.1738338470458984s
Received healthy response to inference request in 2.4142487049102783s
Received healthy response to inference request in 2.429753065109253s
Received healthy response to inference request in 2.3555102348327637s
5 requests
0 failed requests
5th percentile: 2.2101691246032713
10th percentile: 2.2465044021606446
20th percentile: 2.319174957275391
30th percentile: 2.3672579288482667
40th percentile: 2.3907533168792723
50th percentile: 2.4142487049102783
60th percentile: 2.420450448989868
70th percentile: 2.426652193069458
80th percentile: 2.5733758926391603
90th percentile: 2.8606215476989747
95th percentile: 3.0042443752288817
99th percentile: 3.1191426372528075
mean time: 2.5042426109313967
Pipeline stage StressChecker completed in 14.42s
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 0.64s
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 0.76s
Shutdown handler de-registered
chaiml-ir-v3-4-dpo-irv0f_3260_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 0.09s
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-ir-v3-4-dpo-irv0f-3260-v1-profiler
Waiting for inference service chaiml-ir-v3-4-dpo-irv0f-3260-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
%s, retrying in %s seconds...
Evaluating %s Family Friendly Score with %s threads
Pipeline stage OfflineFamilyFriendlyScorer completed in 5161.44s
Shutdown handler de-registered
chaiml-ir-v3-4-dpo-irv0f_3260_v1 status is now torndown due to DeploymentManager action