developer_uid: chai_backend_admin
submission_id: chaiml-ir-v4-2-dpo-ir31_10551_v1
model_name: training123
model_group: ChaiML/ir-v4-2-dpo_ir31c
status: torndown
timestamp: 2025-11-07T09:51:18+00:00
num_battles: 5586
num_wins: 2908
celo_rating: 1303.86
family_friendly_score: 0.44220000000000004
family_friendly_standard_error: 0.007023662292565041
submission_type: basic
model_repo: ChaiML/ir-v4-2-dpo_ir31cos-lr5e6b01-cosine
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': 0.3826878796041192, 'latency_mean': 2.612994141578674, 'latency_p50': 2.6054521799087524, 'latency_p90': 2.871423029899597}, {'batch_size': 2, 'throughput': 0.579562848594905, 'latency_mean': 3.442605230808258, 'latency_p50': 3.4353963136672974, 'latency_p90': 3.7452489614486693}, {'batch_size': 3, 'throughput': 0.7069833128347474, 'latency_mean': 4.239509938955307, 'latency_p50': 4.267956733703613, 'latency_p90': 4.560949444770813}, {'batch_size': 4, 'throughput': 0.809846259881064, 'latency_mean': 4.926983292102814, 'latency_p50': 4.923572540283203, 'latency_p90': 5.60259895324707}, {'batch_size': 5, 'throughput': 0.8757772647618297, 'latency_mean': 5.688067661523819, 'latency_p50': 5.667624115943909, 'latency_p90': 6.508524775505066}]
gpu_counts: {'NVIDIA L40S': 1}
display_name: training123
is_internal_developer: True
language_model: ChaiML/ir-v4-2-dpo_ir31cos-lr5e6b01-cosine
model_size: 24B
ranking_group: single
throughput_3p7s: 0.63
us_pacific_date: 2025-11-04
win_ratio: 0.5205871822413176
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:', '</s>', '\n'], '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
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
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Running pipeline stage MKMLizer
Starting job with name chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer
Waiting for job on chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer to finish
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: ║                                                                     ║
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: ║   ██████   ██████  █████   ████  ████                               ║
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: ║  ░░██████ ██████  ░░███   ███░  ░░███                               ║
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: ║   ░███░█████░███   ░███  ███     ░███                               ║
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: ║   ░███░░███ ░███   ░███████      ░███                               ║
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: ║   ░███ ░░░  ░███   ░███░░███     ░███                               ║
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: ║   ░███      ░███   ░███ ░░███    ░███                               ║
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: ║   █████     █████  █████ ░░████  █████                              ║
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: ║  ░░░░░     ░░░░░  ░░░░░   ░░░░  ░░░░░                               ║
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: ║                                                                     ║
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: ║  Version: 0.30.2                                                    ║
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: ║  Features: FLYWHEEL, CUDA                                           ║
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: ║  Copyright 2023-2025 MK ONE TECHNOLOGIES Inc.                       ║
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: ║  https://mk1.ai                                                     ║
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: ║                                                                     ║
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: ║  belonging to:                                                      ║
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: ║                                                                     ║
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: ║  Chai Research Corp.                                                ║
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: ║  Expiration: 2028-03-31 23:59:59                                    ║
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: ║                                                                     ║
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
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-v4-2-dpo-ir31-10551-v1-mkmlizer: Downloaded to shared memory in 56.340s
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: Checking if ChaiML/ir-v4-2-dpo_ir31cos-lr5e6b01-cosine already exists in ChaiML
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpk94o1exb, device:0
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: quantized model in 42.834s
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: Processed model ChaiML/ir-v4-2-dpo_ir31cos-lr5e6b01-cosine in 99.174s
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: creating bucket guanaco-mkml-models
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-ir-v4-2-dpo-ir31-10551-v1/nvidia
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-ir-v4-2-dpo-ir31-10551-v1/nvidia/config.json
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-ir-v4-2-dpo-ir31-10551-v1/nvidia/special_tokens_map.json
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-ir-v4-2-dpo-ir31-10551-v1/nvidia/tokenizer_config.json
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-ir-v4-2-dpo-ir31-10551-v1/nvidia/tokenizer.json
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-ir-v4-2-dpo-ir31-10551-v1/nvidia/flywheel_model.0.safetensors
chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/chaiml-ir-v4-2-dpo-ir31-10551-v1/nvidia/flywheel_model.1.safetensors
Job chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer completed after 154.22s with status: succeeded
Stopping job with name chaiml-ir-v4-2-dpo-ir31-10551-v1-mkmlizer
Pipeline stage MKMLizer completed in 154.84s
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-v4-2-dpo-ir31-10551-v1
Waiting for inference service chaiml-ir-v4-2-dpo-ir31-10551-v1 to be ready
Inference service chaiml-ir-v4-2-dpo-ir31-10551-v1 ready after 140.91212940216064s
Pipeline stage MKMLDeployer completed in 141.50s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.505228281021118s
Received healthy response to inference request in 2.6519384384155273s
Received healthy response to inference request in 2.317166566848755s
Received healthy response to inference request in 2.375715970993042s
Received healthy response to inference request in 2.288900136947632s
5 requests
0 failed requests
5th percentile: 2.2945534229278564
10th percentile: 2.300206708908081
20th percentile: 2.3115132808685304
30th percentile: 2.3288764476776125
40th percentile: 2.3522962093353272
50th percentile: 2.375715970993042
60th percentile: 2.4275208950042724
70th percentile: 2.4793258190155028
80th percentile: 2.5345703125
90th percentile: 2.5932543754577635
95th percentile: 2.6225964069366454
99th percentile: 2.646070032119751
mean time: 2.427789878845215
Pipeline stage StressChecker completed in 14.20s
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 0.67s
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 1.00s
Shutdown handler de-registered
chaiml-ir-v4-2-dpo-ir31_10551_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 0.09s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service chaiml-ir-v4-2-dpo-ir31-10551-v1-profiler
Waiting for inference service chaiml-ir-v4-2-dpo-ir31-10551-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
Pipeline stage OfflineFamilyFriendlyScorer completed in 3074.60s
Shutdown handler de-registered
chaiml-ir-v4-2-dpo-ir31_10551_v1 status is now inactive due to auto deactivation removed underperforming models
chaiml-ir-v4-2-dpo-ir31_10551_v1 status is now torndown due to DeploymentManager action