chaiml-kimid-v4a

developer_uid: chai_backend_admin

submission_id: chaiml-kimid-v4a_v1

model_name: training123

model_group: ChaiML/kimid-v4a

status: inactive

timestamp: 2025-12-06T15:49:28+00:00

num_battles: 6650

num_wins: 3438

celo_rating: 1305.48

family_friendly_score: 0.49939999999999996

family_friendly_standard_error: 0.0070710627206948175

submission_type: basic

model_repo: ChaiML/kimid-v4a

model_architecture: MistralForCausalLM

model_num_parameters: 24096691200.0

best_of: 8

max_input_tokens: 2048

max_output_tokens: 72

reward_model: default

latencies: [{'batch_size': 1, 'throughput': 0.34694059109108705, 'latency_mean': 2.882260981798172, 'latency_p50': 2.864413022994995, 'latency_p90': 3.163583993911743}, {'batch_size': 2, 'throughput': 0.5247066778877203, 'latency_mean': 3.811003556251526, 'latency_p50': 3.8067275285720825, 'latency_p90': 4.153888845443726}, {'batch_size': 3, 'throughput': 0.6486559993665716, 'latency_mean': 4.612573816776275, 'latency_p50': 4.622984051704407, 'latency_p90': 5.176049208641052}, {'batch_size': 4, 'throughput': 0.7299200373474787, 'latency_mean': 5.462292072772979, 'latency_p50': 5.491538763046265, 'latency_p90': 6.043499302864075}, {'batch_size': 5, 'throughput': 0.794525244942463, 'latency_mean': 6.265177595615387, 'latency_p50': 6.298694610595703, 'latency_p90': 7.101075434684754}]

gpu_counts: {'NVIDIA L40S': 1}

display_name: training123

ineligible_reason: max_output_tokens!=64

is_internal_developer: True

language_model: ChaiML/kimid-v4a

model_size: 24B

ranking_group: single

throughput_3p7s: 0.51

us_pacific_date: 2025-12-06

win_ratio: 0.5169924812030076

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:', '</s>', 'User:'], 'max_input_tokens': 2048, 'best_of': 8, 'max_output_tokens': 72}

formatter: {'memory_template': "{bot_name}'s Persona: {memory}\n####\n", 'prompt_template': '', 'bot_template': '{bot_name}: {message}\n', 'user_template': '{user_name}: {message}\n', 'response_template': '####\n{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-kimid-v4a-v1-mkmlizer
Waiting for job on chaiml-kimid-v4a-v1-mkmlizer to finish
chaiml-kimid-v4a-v1-mkmlizer: bash: cannot set terminal process group (-1): Inappropriate ioctl for device
chaiml-kimid-v4a-v1-mkmlizer: bash: no job control in this shell
chaiml-kimid-v4a-v1-mkmlizer: /root/miniconda3/envs/nvidia/lib/python3.11/site-packages/mk1/__init__.py:1: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.
chaiml-kimid-v4a-v1-mkmlizer:   __import__('pkg_resources').declare_namespace(__name__)
chaiml-kimid-v4a-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-kimid-v4a-v1-mkmlizer: ║                                                                     ║
chaiml-kimid-v4a-v1-mkmlizer: ║   ██████   ██████  █████   ████  ████                               ║
chaiml-kimid-v4a-v1-mkmlizer: ║  ░░██████ ██████  ░░███   ███░  ░░███                               ║
chaiml-kimid-v4a-v1-mkmlizer: ║   ░███░█████░███   ░███  ███     ░███                               ║
chaiml-kimid-v4a-v1-mkmlizer: ║   ░███░░███ ░███   ░███████      ░███                               ║
chaiml-kimid-v4a-v1-mkmlizer: ║   ░███ ░░░  ░███   ░███░░███     ░███                               ║
chaiml-kimid-v4a-v1-mkmlizer: ║   ░███      ░███   ░███ ░░███    ░███                               ║
chaiml-kimid-v4a-v1-mkmlizer: ║   █████     █████  █████ ░░████  █████                              ║
chaiml-kimid-v4a-v1-mkmlizer: ║  ░░░░░     ░░░░░  ░░░░░   ░░░░  ░░░░░                               ║
chaiml-kimid-v4a-v1-mkmlizer: ║                                                                     ║
chaiml-kimid-v4a-v1-mkmlizer: ║  Version: 0.30.6+torch280                                           ║
chaiml-kimid-v4a-v1-mkmlizer: ║  Features: FLYWHEEL, CUDA                                           ║
chaiml-kimid-v4a-v1-mkmlizer: ║  Copyright 2023-2025 MK ONE TECHNOLOGIES Inc.                       ║
chaiml-kimid-v4a-v1-mkmlizer: ║  https://mk1.ai                                                     ║
chaiml-kimid-v4a-v1-mkmlizer: ║                                                                     ║
chaiml-kimid-v4a-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
chaiml-kimid-v4a-v1-mkmlizer: ║  belonging to:                                                      ║
chaiml-kimid-v4a-v1-mkmlizer: ║                                                                     ║
chaiml-kimid-v4a-v1-mkmlizer: ║  Chai Research Corp.                                                ║
chaiml-kimid-v4a-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
chaiml-kimid-v4a-v1-mkmlizer: ║  Expiration: 2028-03-31 23:59:59                                    ║
chaiml-kimid-v4a-v1-mkmlizer: ║                                                                     ║
chaiml-kimid-v4a-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
chaiml-kimid-v4a-v1-mkmlizer: Downloaded to shared memory in 80.066s
chaiml-kimid-v4a-v1-mkmlizer: Checking if ChaiML/kimid-v4a already exists in ChaiML
chaiml-kimid-v4a-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp9egvi6s9, device:0
chaiml-kimid-v4a-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
chaiml-kimid-v4a-v1-mkmlizer: 
Loading 0:   0%|          | 0.00/363 [00:00<?, ?it/s]
Loading 0:   9%|▉         | 33.0/363 [00:01<00:11, 29.8it/s]
Loading 0:   9%|▉         | 33.0/363 [00:01<00:11, 29.8it/s]
Loading 0:  16%|█▋        | 59.0/363 [00:02<00:11, 27.5it/s]
Loading 0:  16%|█▋        | 59.0/363 [00:02<00:11, 27.5it/s]
Loading 0:  23%|██▎       | 84.0/363 [00:03<00:10, 26.3it/s]
Loading 0:  23%|██▎       | 84.0/363 [00:03<00:10, 26.3it/s]
Loading 0:  31%|███       | 111/363 [00:04<00:09, 26.5it/s] 
Loading 0:  31%|███       | 111/363 [00:04<00:09, 26.5it/s]
Loading 0:  39%|███▊      | 140/363 [00:05<00:08, 26.5it/s]
Loading 0:  39%|███▊      | 140/363 [00:05<00:08, 26.5it/s]
Loading 0:  46%|████▌     | 167/363 [00:06<00:07, 26.2it/s]
Loading 0:  46%|████▌     | 167/363 [00:06<00:07, 26.2it/s]
Loading 0:  53%|█████▎    | 194/363 [00:07<00:06, 26.0it/s]
Loading 0:  53%|█████▎    | 194/363 [00:07<00:06, 26.0it/s]
Loading 0:  53%|█████▎    | 194/363 [00:20<00:06, 26.0it/s]
Loading 0:  55%|█████▌    | 201/363 [00:20<00:37, 4.27it/s]
Loading 0:  55%|█████▌    | 201/363 [00:20<00:37, 4.27it/s]
Loading 0:  62%|██████▏   | 224/363 [00:21<00:23, 5.81it/s]
Loading 0:  62%|██████▏   | 224/363 [00:21<00:23, 5.81it/s]
Loading 0:  70%|███████   | 255/363 [00:22<00:12, 8.48it/s]
Loading 0:  70%|███████   | 255/363 [00:22<00:12, 8.48it/s]
Loading 0:  77%|███████▋  | 280/363 [00:23<00:07, 10.6it/s]
Loading 0:  77%|███████▋  | 280/363 [00:23<00:07, 10.6it/s]
Loading 0:  84%|████████▍ | 306/363 [00:24<00:04, 12.9it/s]
Loading 0:  84%|████████▍ | 306/363 [00:24<00:04, 12.9it/s]
Loading 0:  93%|█████████▎| 336/363 [00:25<00:01, 15.7it/s]
Loading 0:  93%|█████████▎| 336/363 [00:25<00:01, 15.7it/s]
Loading 0:  99%|█████████▉| 360/363 [00:26<00:00, 17.4it/s]
Loading 0:  99%|█████████▉| 360/363 [00:26<00:00, 17.4it/s]
Loading 0: 100%|██████████| 363/363 [00:26<00:00, 17.5it/s]
Loading 0: 100%|██████████| 363/363 [00:26<00:00, 17.5it/s]
                                                           

Loading 0: 100%|██████████| 363/363 [00:26<00:00, 13.5it/s]
chaiml-kimid-v4a-v1-mkmlizer: The tokenizer you are loading from '/tmp/tmp9egvi6s9' with an incorrect regex pattern: https://huggingface.co/mistralai/Mistral-Small-3.1-24B-Instruct-2503/discussions/84#69121093e8b480e709447d5e. This will lead to incorrect tokenization. You should set the `fix_mistral_regex=True` flag when loading this tokenizer to fix this issue.
chaiml-kimid-v4a-v1-mkmlizer: quantized model in 43.764s
chaiml-kimid-v4a-v1-mkmlizer: Processed model ChaiML/kimid-v4a in 123.830s
chaiml-kimid-v4a-v1-mkmlizer: creating bucket guanaco-mkml-models
chaiml-kimid-v4a-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-kimid-v4a-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-kimid-v4a-v1/nvidia
chaiml-kimid-v4a-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-kimid-v4a-v1/nvidia/config.json
chaiml-kimid-v4a-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-kimid-v4a-v1/nvidia/special_tokens_map.json
chaiml-kimid-v4a-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-kimid-v4a-v1/nvidia/tokenizer_config.json
chaiml-kimid-v4a-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/chaiml-kimid-v4a-v1/nvidia/flywheel_model.1.safetensors
chaiml-kimid-v4a-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-kimid-v4a-v1/nvidia/flywheel_model.0.safetensors
Job chaiml-kimid-v4a-v1-mkmlizer completed after 196.58s with status: succeeded
Stopping job with name chaiml-kimid-v4a-v1-mkmlizer
Pipeline stage MKMLizer completed in 197.06s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.17s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service chaiml-kimid-v4a-v1
Waiting for inference service chaiml-kimid-v4a-v1 to be ready
Inference service chaiml-kimid-v4a-v1 ready after 191.32420325279236s
Pipeline stage MKMLDeployer completed in 192.22s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 3.613252878189087s
Received healthy response to inference request in 3.5396006107330322s
Received healthy response to inference request in 3.4151222705841064s
Received healthy response to inference request in 3.003004550933838s
Received healthy response to inference request in 3.057441473007202s
5 requests
0 failed requests
5th percentile: 3.0138919353485107
10th percentile: 3.0247793197631836
20th percentile: 3.0465540885925293
30th percentile: 3.128977632522583
40th percentile: 3.2720499515533445
50th percentile: 3.4151222705841064
60th percentile: 3.4649136066436768
70th percentile: 3.514704942703247
80th percentile: 3.554331064224243
90th percentile: 3.583791971206665
95th percentile: 3.598522424697876
99th percentile: 3.6103067874908445
mean time: 3.325684356689453
Pipeline stage StressChecker completed in 22.79s
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 2.06s
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 2.71s
Shutdown handler de-registered
chaiml-kimid-v4a_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
Generating Leaderboard row for %s
Generated Leaderboard row for %s
Pipeline stage OfflineFamilyFriendlyScorer completed in 2928.92s
Shutdown handler de-registered