developer_uid: chai_backend_admin
submission_id: chaiml-icld-v4-dpo_v1
model_name: training123
model_group: ChaiML/icld-v4-dpo
status: protected
timestamp: 2025-11-25T00:25:02+00:00
num_battles: 5891
num_wins: 3079
celo_rating: 1315.26
family_friendly_score: 0.5329999999999999
family_friendly_standard_error: 0.007055650218087628
submission_type: basic
model_repo: ChaiML/icld-v4-dpo
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': 0.2118143037380136, 'latency_mean': 4.720991957187652, 'latency_p50': 4.536151170730591, 'latency_p90': 5.982544302940369}, {'batch_size': 2, 'throughput': 0.5759448701257057, 'latency_mean': 3.4592540037631987, 'latency_p50': 3.4718674421310425, 'latency_p90': 3.7815361976623536}, {'batch_size': 3, 'throughput': 0.7150200703042562, 'latency_mean': 4.184743499755859, 'latency_p50': 4.175861239433289, 'latency_p90': 4.569168424606323}, {'batch_size': 4, 'throughput': 0.8062152584479528, 'latency_mean': 4.953746845722199, 'latency_p50': 4.954337239265442, 'latency_p90': 5.606547617912292}, {'batch_size': 5, 'throughput': 0.678973603916594, 'latency_mean': 7.339114396572113, 'latency_p50': 7.170686602592468, 'latency_p90': 9.81017508506775}]
gpu_counts: {'NVIDIA L40S': 1}
display_name: training123
is_internal_developer: True
language_model: ChaiML/icld-v4-dpo
model_size: 24B
ranking_group: single
throughput_3p7s: 0.27
us_pacific_date: 2025-11-23
win_ratio: 0.5226616873196401
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': '', '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-icld-v4-dpo-v1-mkmlizer
Waiting for job on chaiml-icld-v4-dpo-v1-mkmlizer to finish
chaiml-icld-v4-dpo-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-icld-v4-dpo-v1-mkmlizer: ║                                                                     ║
chaiml-icld-v4-dpo-v1-mkmlizer: ║   ██████   ██████  █████   ████  ████                               ║
chaiml-icld-v4-dpo-v1-mkmlizer: ║  ░░██████ ██████  ░░███   ███░  ░░███                               ║
chaiml-icld-v4-dpo-v1-mkmlizer: ║   ░███░█████░███   ░███  ███     ░███                               ║
chaiml-icld-v4-dpo-v1-mkmlizer: ║   ░███░░███ ░███   ░███████      ░███                               ║
chaiml-icld-v4-dpo-v1-mkmlizer: ║   ░███ ░░░  ░███   ░███░░███     ░███                               ║
chaiml-icld-v4-dpo-v1-mkmlizer: ║   ░███      ░███   ░███ ░░███    ░███                               ║
chaiml-icld-v4-dpo-v1-mkmlizer: ║   █████     █████  █████ ░░████  █████                              ║
chaiml-icld-v4-dpo-v1-mkmlizer: ║  ░░░░░     ░░░░░  ░░░░░   ░░░░  ░░░░░                               ║
chaiml-icld-v4-dpo-v1-mkmlizer: ║                                                                     ║
chaiml-icld-v4-dpo-v1-mkmlizer: ║  Version: 0.30.2                                                    ║
chaiml-icld-v4-dpo-v1-mkmlizer: ║  Features: FLYWHEEL, CUDA                                           ║
chaiml-icld-v4-dpo-v1-mkmlizer: ║  Copyright 2023-2025 MK ONE TECHNOLOGIES Inc.                       ║
chaiml-icld-v4-dpo-v1-mkmlizer: ║  https://mk1.ai                                                     ║
chaiml-icld-v4-dpo-v1-mkmlizer: ║                                                                     ║
chaiml-icld-v4-dpo-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
chaiml-icld-v4-dpo-v1-mkmlizer: ║  belonging to:                                                      ║
chaiml-icld-v4-dpo-v1-mkmlizer: ║                                                                     ║
chaiml-icld-v4-dpo-v1-mkmlizer: ║  Chai Research Corp.                                                ║
chaiml-icld-v4-dpo-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
chaiml-icld-v4-dpo-v1-mkmlizer: ║  Expiration: 2028-03-31 23:59:59                                    ║
chaiml-icld-v4-dpo-v1-mkmlizer: ║                                                                     ║
chaiml-icld-v4-dpo-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
chaiml-icld-v4-dpo-v1-mkmlizer: Downloaded to shared memory in 100.323s
chaiml-icld-v4-dpo-v1-mkmlizer: Checking if ChaiML/icld-v4-dpo already exists in ChaiML
chaiml-icld-v4-dpo-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpr6pdi58_, device:0
chaiml-icld-v4-dpo-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
chaiml-icld-v4-dpo-v1-mkmlizer: quantized model in 47.779s
chaiml-icld-v4-dpo-v1-mkmlizer: Processed model ChaiML/icld-v4-dpo in 148.102s
chaiml-icld-v4-dpo-v1-mkmlizer: creating bucket guanaco-mkml-models
chaiml-icld-v4-dpo-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-icld-v4-dpo-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-icld-v4-dpo-v1/nvidia
chaiml-icld-v4-dpo-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-icld-v4-dpo-v1/nvidia/config.json
chaiml-icld-v4-dpo-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-icld-v4-dpo-v1/nvidia/special_tokens_map.json
chaiml-icld-v4-dpo-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-icld-v4-dpo-v1/nvidia/tokenizer.json
chaiml-icld-v4-dpo-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-icld-v4-dpo-v1/nvidia/flywheel_model.0.safetensors
chaiml-icld-v4-dpo-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/chaiml-icld-v4-dpo-v1/nvidia/flywheel_model.1.safetensors
Job chaiml-icld-v4-dpo-v1-mkmlizer completed after 195.61s with status: succeeded
Stopping job with name chaiml-icld-v4-dpo-v1-mkmlizer
Pipeline stage MKMLizer completed in 196.31s
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-icld-v4-dpo-v1
Waiting for inference service chaiml-icld-v4-dpo-v1 to be ready
Inference service chaiml-icld-v4-dpo-v1 ready after 141.0053849220276s
Pipeline stage MKMLDeployer completed in 142.07s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.666529417037964s
Received healthy response to inference request in 2.4610328674316406s
Received healthy response to inference request in 2.0804929733276367s
Received healthy response to inference request in 2.07231068611145s
Received healthy response to inference request in 2.1320395469665527s
5 requests
0 failed requests
5th percentile: 2.0739471435546877
10th percentile: 2.0755836009979247
20th percentile: 2.0788565158843992
30th percentile: 2.09080228805542
40th percentile: 2.111420917510986
50th percentile: 2.1320395469665527
60th percentile: 2.2636368751525877
70th percentile: 2.395234203338623
80th percentile: 2.5021321773529053
90th percentile: 2.5843307971954346
95th percentile: 2.625430107116699
99th percentile: 2.658309555053711
mean time: 2.2824810981750487
Pipeline stage StressChecker completed in 12.99s
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 0.56s
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-icld-v4-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
Retrying (%r) after connection broken by '%r': %s
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 0.07s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service chaiml-icld-v4-dpo-v1-profiler
Waiting for inference service chaiml-icld-v4-dpo-v1-profiler to be ready
Inference service chaiml-icld-v4-dpo-v1-profiler ready after 151.1219425201416s
Pipeline stage MKMLProfilerDeployer completed in 151.81s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl --kubeconfig /code/guanaco/guanaco_services/resources/kchai_coreweave_us_east_04a.yaml cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/chaiml-icld-v4-dpo-v1-profiler-predictor-00001-deployment-wzjnw:/code/chaiverse_profiler_1763957838 --namespace tenant-chaiml-guanaco
kubectl --kubeconfig /code/guanaco/guanaco_services/resources/kchai_coreweave_us_east_04a.yaml exec -it chaiml-icld-v4-dpo-v1-profiler-predictor-00001-deployment-wzjnw --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1763957838 && python profiles.py profile --best_of_n 8 --auto_batch 5 --batches 1,5,10,15,20,25,30,35,40,45,50,55,60,65,70,75,80,85,90,95,100,105,110,115,120,125,130,135,140,145,150,155,160,165,170,175,180,185,190,195 --samples 200 --input_tokens 2048 --output_tokens 64 --summary /code/chaiverse_profiler_1763957838/summary.json'
%s, retrying in %s seconds...
kubectl --kubeconfig /code/guanaco/guanaco_services/resources/kchai_coreweave_us_east_04a.yaml cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/chaiml-icld-v4-dpo-v1-profiler-predictor-00001-deployment-wzjnw:/code/chaiverse_profiler_1763958573 --namespace tenant-chaiml-guanaco
kubectl --kubeconfig /code/guanaco/guanaco_services/resources/kchai_coreweave_us_east_04a.yaml exec -it chaiml-icld-v4-dpo-v1-profiler-predictor-00001-deployment-wzjnw --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1763958573 && python profiles.py profile --best_of_n 8 --auto_batch 5 --batches 1,5,10,15,20,25,30,35,40,45,50,55,60,65,70,75,80,85,90,95,100,105,110,115,120,125,130,135,140,145,150,155,160,165,170,175,180,185,190,195 --samples 200 --input_tokens 2048 --output_tokens 64 --summary /code/chaiverse_profiler_1763958573/summary.json'
kubectl --kubeconfig /code/guanaco/guanaco_services/resources/kchai_coreweave_us_east_04a.yaml exec -it chaiml-icld-v4-dpo-v1-profiler-predictor-00001-deployment-wzjnw --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1763958573/summary.json'
Pipeline stage MKMLProfilerRunner completed in 2853.86s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service chaiml-icld-v4-dpo-v1-profiler is running
Tearing down inference service chaiml-icld-v4-dpo-v1-profiler
Service chaiml-icld-v4-dpo-v1-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 0.77s
Shutdown handler de-registered
chaiml-icld-v4-dpo_v1 status is now inactive due to auto deactivation removed underperforming models
chaiml-icld-v4-dpo_v1 status is now protected due to ABTestQueueItem
chaiml-icld-v4-dpo_v1 status is now protected due to ABTestQueueItem
chaiml-icld-v4-dpo_v1 status is now protected due to ABTestQueueItem
chaiml-icld-v4-dpo_v1 status is now protected due to ABTestQueueItem